affaan-m/ECC

551 skills · 3M installs · 491k stars · GitHub

Install

npx skills add https://github.com/affaan-m/ECC

Skills in this repo

1Security Reviewsecurity-review is an ECC-origin agent skill that activates when developers implement authentication, authorization, user input, file uploads, API endpoints, secrets, payments, or third-party integrations. The skill walks through security best practices and identifies common vulnerability classes before code merges or deploys. Developers reach for security-review when sensitive features land and they need a structured audit instead of ad-hoc grep passes. Triggers cover storing or transmitting sensitive data and integrating external APIs. The checklist spans auth flows, input validation, credential handling, and payment surfaces so gaps surface early in the review cycle.18kinstalls 2Golang Patternsgolang-patterns is an affaan-m/ecc skill activated on **/*.go, **/go.mod, and **/go.sum files, extending design principles with Go-specific idioms. Coverage includes functional options constructors, small interface design, dependency injection, concurrency patterns, error handling discipline, and package organization. Example snippets show Option func(*Server) helpers like WithPort for flexible configuration without sprawling constructors. Developers reach for golang-patterns when refactoring Go microservices, CLIs, or libraries to avoid fighting language quirks and to align new modules with community-accepted structure before code review.15.4kinstalls 3Backend Patternsbackend-patterns is an affaan-m/ecc skill cataloging server-side architecture guidance for Node.js, Express, and Next.js API routes. It activates when designing REST or GraphQL endpoints, implementing repository-service-controller layers, optimizing database queries against N+1 problems with indexing and connection pooling, adding Redis or in-memory caching with HTTP cache headers, setting up background jobs, and structuring validation plus middleware for auth, logging, and rate limiting. The skill encodes repeatable patterns for scalable APIs rather than one-off snippets. Developers reach for backend-patterns when scaffolding a new service layer, hardening error handling, or reviewing data-access performance before production load.15kinstalls 4Frontend Patternsfrontend-patterns is an ecc repository skill documenting modern frontend development patterns for React, Next.js, and performant user interfaces. It activates when building components with composition and props, managing state through useState, useReducer, Zustand, or Context, implementing data fetching with SWR, React Query, or server components, optimizing via memoization and virtualization, and handling forms with Zod validation plus accessible responsive UI. Developers reach for frontend-patterns when an agent might otherwise guess at state, fetching, or performance tradeoffs during everyday React and Next.js feature work across multiple frontend concerns.14.8kinstalls 5Coding Standardscoding-standards is a cross-project Claude Code skill from affaan-m/ecc that sets the minimum bar for naming, immutability, readability, and code-quality review before agents reach for specialized playbooks. The skill explicitly defers React, state, and UI architecture to frontend-patterns, repository and endpoint design to backend-patterns or api-design, and the shortest reusable rule layer to rules/common/coding-style.md. Developers reach for coding-standards when onboarding an agent onto a new repo, standardizing PR review language, or preventing inconsistent style drift across services written in different frameworks. It is the shared floor, not the detailed framework playbook, so teams pair it with domain skills rather than replacing them.14.7kinstalls 6Golang Testinggolang-testing is an ECC-origin skill that teaches and generates Go testing patterns aligned with the standard go test toolchain. It centers on table-driven tests as the primary idiom, with guidance for test helpers, benchmarking, race detection, coverage analysis, integration testing, and fuzzing. The skill activates on Go source files matched by globs for **/*.go, go.mod, and go.sum, making it repo-aware during test authoring sessions. Developers reach for golang-testing when writing new Test functions, improving flaky suites, or applying TDD cycles in Go services and CLIs. Patterns extend general testing principles with Go-specific conventions such as subtests, parallel flags, and helper hygiene so generated tests read like idiomatic production code rather than generic templates.13.7kinstalls 7Docker Patternsdocker-patterns is an ECC agent skill cataloging Docker and Docker Compose best practices for containerized development. It guides developers through local Compose setups, multi-container architectures, networking and volume strategies, and Dockerfile reviews for image size and security. Reach for docker-patterns when migrating from bare-metal local dev to containers, troubleshooting service discovery or persistent storage, or standardizing repeatable dev environments across a team. The skill emphasizes practical patterns over theory: compose file structure, inter-service networking, volume mounts, and secure base-image choices. It suits backend and full-stack engineers standing up dependent services like databases, queues, and APIs behind a single docker compose up workflow.13.7kinstalls 8Video Editingvideo-editing is a Claude Code skill from affaan-m/ecc for AI-assisted editing of real footage—not prompt-based generation. The pipeline spans raw capture through FFmpeg cuts, Remotion compositions, ElevenLabs voiceover, fal.ai augmentations, and final polish in Descript or CapCut. Developers reach for video-editing when turning long recordings into short-form content, building tutorials or demo videos, or adding overlays, subtitles, and music to screen captures. The skill activates for vlog structuring, footage cutting, and platform-specific clip preparation from existing video assets.13.4kinstalls 9Springboot Patternsspringboot-patterns is an ECC skill for developers building scalable Spring Boot backends with consistent architecture. It activates when creating REST APIs with Spring MVC or WebFlux, structuring controller-to-service-to-repository layers, configuring Spring Data JPA, caching, or async processing, and adding validation, exception handling, or pagination. The skill also covers profile-based dev, staging, and production configuration plus event-driven patterns with Spring Events. Developers reach for springboot-patterns when a Java API needs opinionated layering and cross-cutting concerns—rather than one-off controller snippets—so new endpoints follow the same validation, logging, and data-access conventions across the codebase.12.9kinstalls 10Python Patternspython-patterns is an ECC-origin agent skill from affaan-m/ecc that activates on Python source via globs for **/*.py and **/*.pyi. It teaches structural subtyping with Protocol, dataclasses, context managers, decorators, async/await, type hints, and package layout patterns drawn from battle-tested practice. When an agent edits or generates Python, the skill steers choices toward duck typing with types, safer resource handling, and clearer module boundaries that reduce bugs and speed maintenance. Developers reach for python-patterns when refactoring legacy scripts, scaffolding APIs, or reviewing Python that feels unidiomatic. It is a build-time companion for anyone shipping Python services, CLIs, or libraries who wants consistent patterns without re-reading style guides on every file.12.8kinstalls 11Frontend Slidesfrontend-slides from affaan-m/ecc builds stunning, animation-rich HTML presentations that run entirely in the browser with zero dependencies. The skill activates for talk decks, pitch decks, workshop decks, or internal presentations, and can convert PPT/PPTX files into web slides. It guides visual exploration so developers pick aesthetics through rendered examples rather than abstract design briefs, inspired by zarazhangrui’s exploration approach. Teams reach for frontend-slides when they need shareable HTML decks for demos or conferences without wiring a slide framework or design toolchain.12.7kinstalls 12Api Designapi-design is an affaan-m/ecc skill that codifies REST API design patterns for consistent, developer-friendly HTTP services. It covers plural kebab-case resource URLs, CRUD route shapes, pagination and filtering query parameters, structured error payloads, versioning strategy, and rate limiting guidance. Developers reach for api-design when designing new endpoints, reviewing existing contracts, or planning public or partner APIs where convention consistency matters. The skill activates on endpoint design, pagination additions, error-handling plans, and versioning discussions rather than implementation code generation alone.12.6kinstalls 13Django Patternsdjango-patterns is an ECC skill that encodes production-grade Django architecture for scalable, maintainable applications. It activates when building Django web apps, designing Django REST Framework APIs, working with the ORM and models, or setting up project structure with caching, signals, and middleware. The skill recommends a split settings layout under config/settings with base.py and environment-specific modules, plus conventions for apps, models, serializers, and API endpoints. Developers reach for django-patterns when scaffolding a new Django service, refactoring a monolithic settings file, or aligning DRF serializers and viewsets with team standards. Guidance stays inside the agent workflow so architecture decisions land directly in generated code and file trees.12.6kinstalls 14Postgres Patternspostgres-patterns is an ECC agent skill offering battle-tested PostgreSQL patterns for schema design, indexing, data types, query optimization, and Row Level Security. The skill includes an index cheat sheet mapping query patterns to index types, anti-pattern detection guidance, and connection pooling notes, credited to Supabase best practices under MIT License. Developers reach for postgres-patterns when writing SQL migrations, troubleshooting slow queries, or implementing RLS policies. For deeper review workflows, the skill points to the database-reviewer agent while serving as an in-editor quick reference during build.12.6kinstalls 15Tdd Workflowtdd-workflow is a TDD enforcement skill activated when writing new features, fixing bugs, refactoring, adding API endpoints, or creating components. Core rules require tests before code and a minimum 80% coverage spanning unit, integration, and end-to-end tests. Developers reach for tdd-workflow when an agent session drifts into implementation-first habits or when teams want consistent red-green-refactor discipline across bug fixes and new endpoints. The skill defines activation triggers and coverage expectations rather than binding to one test framework, so it pairs with project-specific runners the agent discovers in the repo.12.2kinstalls 16Prompt Optimizerprompt-optimizer is an ECC (Everything Claude Code) skill that turns vague or weak prompts into precise, ECC-aware instructions Claude Code or Cursor can execute reliably. It targets developers who struggle with ambiguous agent requests that produce inconsistent edits, missed constraints, or partial implementations. The skill aligns phrasing with ECC conventions so downstream agents inherit clear scope, constraints, and success criteria. Reach for prompt-optimizer before kicking off complex multi-file changes, refactors, or workflow automations where prompt quality directly affects output. The repository notes a Japanese translation is in progress; the English ECC origin skill defines the authoritative behavior.11.9kinstalls 17Java Coding Standardsjava-coding-standards is an ECC agent skill that enforces readable, maintainable Java 17+ conventions across Spring Boot and Quarkus microservices. The skill automatically applies framework-specific rules for naming, immutability, Optional usage, streams, exception handling, generics, CDI, reactive patterns, and project package layout. It activates when writing or reviewing Java backend code, structuring records and sealed classes, applying pattern matching, or aligning Quarkus and Spring Boot services to shared ECC standards. java-coding-standards covers Optional and stream idioms, exception boundaries, and reactive service patterns so agents produce consistent JVM code instead of ad hoc style per file. Developers reach for java-coding-standards when onboarding agents onto existing Spring Boot or Quarkus repos that need enforced conventions during implementation and review.11.9kinstalls 18Django Securitydjango-security is an ECC agent skill with comprehensive Django security guidelines covering authentication, authorization, CSRF protection, SQL injection prevention, XSS prevention, and production settings configuration. The skill walks through production settings.py patterns, user permissions, role-based access, and pre-deployment security reviews. Developers reach for django-security when setting up Django auth, implementing permissions, reviewing apps for vulnerabilities, or configuring secure production deployments. The guidance targets common OWASP-class risks in Django ORM queries, templates, middleware, and environment-specific settings before traffic hits production.11.8kinstalls 19Market Researchmarket-research is a decision-oriented research skill from affaan-m/ecc that turns vague market questions into attributed intelligence developers can act on. The skill guides agents through TAM/SAM/SOM estimation, competitor and adjacent-product comparisons, fund and investor dossiers, and technology trend scans, emphasizing sourced claims over research theater. Developers reach for market-research when pressure-testing a thesis before building, fundraising, or entering a new category, and when they need structured summaries that support go/no-go calls rather than open-ended browsing. It activates on triggers like market sizing, competitive analysis, investor outreach prep, and industry intelligence that informs product or business decisions.11.8kinstalls 20Article Writingarticle-writing is a long-form content skill for developers and technical founders publishing blog posts, essays, guides, tutorials, and newsletter issues. It activates when output must exceed a paragraph and voice consistency, structure, and credibility matter. The skill transforms notes, transcripts, or research into polished drafts that match supplied examples or brand guidance instead of generic LLM tone. Developers reach for article-writing when documenting launches, explaining architecture decisions publicly, or converting internal research into SEO-friendly technical articles without sounding like auto-generated paste.11.8kinstalls 21Python Testingpython-testing is an ECC agent skill for comprehensive Python testing using pytest as the primary framework. The skill documents basic test structure, fixtures, parametrization, mocking dependencies, async testing patterns, coverage analysis, and test organization with globs targeting **/*.py and **/*.pyi files. Developers reach for python-testing when writing new tests, improving coverage toward 80%+, or applying TDD during Python service development. The skill emphasizes clean pytest syntax and maintainable test suites over ad hoc unittest patterns.11.8kinstalls 22Android Clean Architectureandroid-clean-architecture from affaan-m/ecc covers Clean Architecture patterns for Android and Kotlin Multiplatform projects. It defines recommended module layouts—app entry, core utilities, domain UseCases, data repositories—and dependency inversion rules across presentation, domain, and data layers. The skill applies when structuring modules, implementing UseCase, Repository, and DataSource patterns, designing cross-layer data flow, or setting up Koin or Hilt dependency injection with Room, SQLDelight, or Ktor in layered architectures. Developers reach for it when agents need explicit guidance on Android/KMP boundaries instead of ad-hoc package organization.11.6kinstalls 23Springboot Securityspringboot-security is an ECC-origin expert review skill for Java Spring Boot applications using Spring Security. It activates when developers add authentication via JWT, OAuth2, or sessions, implement authorization with @PreAuthorize and role-based access, validate input with Bean Validation, configure CORS and CSRF protections, set security headers, manage secrets through Vault or environment variables, add rate limiting or brute-force defenses, and scan dependencies for known vulnerabilities. The skill provides opinionated best practices across the full request lifecycle so endpoints resist common OWASP-class failures before production deploys. Teams reach for springboot-security during endpoint creation, auth refactors, or pre-release security passes on Spring Boot microservices and monoliths rather than generic Java code review without a security lens.11.6kinstalls 24Cpp Coding Standardscpp-coding-standards is an ECC agent skill derived from the C++ Core Guidelines (isocpp.github.io) for modern C++17, C++20, and C++23. The skill steers AI agents toward type-safe, resource-safe, immutable, and clear patterns when creating classes, functions, templates, or architectural decisions. Developers reach for cpp-coding-standards during new feature work, code review, and refactors where legacy C++ risks undefined behavior, leaks, or inconsistent style. It complements compiler warnings by encoding guideline-level rules agents can apply consistently across files and modules.11.6kinstalls 25Laravel Securitylaravel-security is an ECC agent skill with a complete Laravel security guide covering authentication, authorization, input validation, CSRF middleware, mass-assignment protection, file upload safety, API endpoint hardening, secrets management, rate limiting, and secure production deployment. Developers activate it when adding Laravel auth, handling user input and uploads, building new API routes, or reinforcing environment configuration. The skill explains how Laravel middleware delivers baseline CSRF and session protections and where developers must add validation, policy checks, and deployment hardening. Reach for laravel-security during new endpoint work or pre-production security reviews on Laravel codebases.11.5kinstalls 26Liquid Glass Designliquid-glass-design is an ECC skill for building Apple's iOS 26 Liquid Glass interface material in native Apple frameworks. The skill walks through blur stacks, reflection layers, and interactive morphing patterns that match the system aesthetic across SwiftUI, UIKit, and WidgetKit targets. Developers reach for liquid-glass-design when shipping iOS 26-era interfaces that need depth, glass translucency, and touch-responsive visual transitions without reverse-engineering WWDC samples from scratch. The published readme notes Japanese translation is still in progress, so the English SKILL.md in the ECC repository remains the authoritative implementation reference for layout, materials, and widget constraints.11.4kinstalls 27Strategic Compactstrategic-compact is an ECC agent skill that coaches when to run Claude Code's manual `/compact` command at logical workflow boundaries instead of relying on automatic compaction. The skill activates when sessions approach context limits above 200K tokens, when switching between unrelated tasks, or after major milestones like finishing research before implementation. It maps compaction points to multi-phase work—research, plan, implement, test—so developers retain decisions, file paths, and constraints that arbitrary auto-compaction would drop. Developers reach for strategic-compact when responses slow down, coherence drops, or they notice the model forgetting earlier agreements mid-session. The workflow is advisory: identify boundary, compact, then continue with a cleaner context window.11.3kinstalls 28Database Migrationsdatabase-migrations is an ECC-origin skill covering safe, reversible database schema changes, data migrations, rollbacks, and zero-downtime deployments. It supports PostgreSQL, MySQL, and common ORMs including Prisma, Drizzle, Django ORM, TypeORM, and golang-migrate. Developers reach for database-migrations when creating or altering tables, adding indexes, running backfills, or planning production schema changes without downtime. Core principles emphasize reversibility and production-safe sequencing for schema and data migration work.11.3kinstalls 29Laravel Patternslaravel-patterns is an ECC-origin agent skill for production Laravel architecture covering routing, controllers, Eloquent ORM, service layers, background queues, events, caching, and API resources with pagination. Developers reach for it when building Laravel web apps or APIs, structuring domain logic away from fat controllers, designing Eloquent relationships, or adding jobs and event-driven workflows. The skill enforces clear boundaries—controllers delegate to services, models stay thin, and API responses use resources with consistent pagination. It activates during greenfield Laravel builds and refactors where maintainable layer separation prevents common Laravel anti-patterns.11.2kinstalls 30E2e TestingE2E Testing is an ECC skill cataloging comprehensive Playwright patterns for stable, fast, and maintainable end-to-end suites. It prescribes tests/e2e folder layout with auth, features, and api spec groups plus shared fixtures and playwright.config.ts at the repo root. Page Object Model examples show encapsulating selectors and navigation in dedicated classes consumed by login, browse, search, and create specs. Guidance also covers CI/CD integration, artifact management for traces and screenshots, and flaky test strategies so pipelines stay green under parallel runs. Reach for E2E Testing when bootstrapping Playwright in a web app, refactoring brittle selectors into POM classes, or hardening GitHub Actions jobs that run headless browsers. Assumes TypeScript and Playwright installed; complements unit testing skills rather than replacing them.11.2kinstalls 31Rust Testingrust-testing is an ECC agent skill for comprehensive Rust test design following test-driven development. It guides agents through identifying target functions, traits, or modules, then writing unit and integration tests, async tests, property-based validation, mocks, and performance benchmarks. Developers invoke rust-testing when shipping new Rust logic, backfilling coverage on legacy crates, or standardizing test style across a workspace. The skill encodes repeatable TDD steps so agents produce maintainable #[test] modules, integration test layouts, and coverage-oriented workflows instead of one-off examples.11.2kinstalls 32Springboot Tddspringboot-tdd is an ECC skill for test-driven development on Spring Boot services with a target of 80%+ unit and integration coverage enforced through JaCoCo. The workflow follows classic TDD: write failing tests first, implement minimal passing code, refactor while green, and verify coverage gates. It applies when adding endpoints, fixing bugs, refactoring, or introducing data-access and security logic. The stack includes JUnit 5 for unit tests, Mockito for mocks, MockMvc for web layer tests, and Testcontainers for integration scenarios. Developers reach for springboot-tdd when they want agent-guided red-green-refactor discipline on Java Spring Boot backends rather than ad hoc test-after coding.10.9kinstalls 33Security Scansecurity-scan is an ECC skill that wraps AgentShield to inspect a project's `.claude/` configuration before commits or onboarding. It scans CLAUDE.md for hardcoded secrets and injection patterns, settings.json for permissive allow lists and missing deny rules, mcp.json for risky MCP servers and embedded env secrets, plus hooks and agent definitions. Triggers include new Claude Code project setup, changes to settings or MCP config, pre-commit review, and periodic hygiene checks. Developers reach for security-scan when they want automated guardrails on agent configuration rather than manual checklist review.10.9kinstalls 34Django Tdddjango-tdd is a test-driven development skill for Django applications using pytest, pytest-django, factory_boy, mocking, coverage reporting, and Django REST Framework. It walks developers through the red-green-refactor cycle: write a failing test for models or views, implement minimal code to pass, then refactor while keeping coverage high. Use django-tdd when creating new Django apps, building DRF APIs, setting up test infrastructure, or testing serializers and authentication flows. The workflow emphasizes factory_boy data setup, isolated unit tests, and API endpoint verification instead of manual QA after features ship.10.9kinstalls 35Codebase Onboardingcodebase-onboarding is an ECC skill that systematically analyzes unfamiliar codebases without reading every file. Phase 1 reconnaissance runs parallel checks: package manifests (package.json, go.mod, Cargo.toml, pyproject.toml, pom.xml), framework fingerprints (next.config, nuxt.config, angular.json, vite.config), and entry-point discovery. The output is a structured onboarding guide with architecture map, key entry points, project conventions, and a ready-to-use CLAUDE.md. Developers reach for codebase-onboarding when opening a repo in Claude Code for the first time, joining a new team, or when asked to explain or onboard to an existing repository.10.9kinstalls 36Deployment PatternsDeployment Patterns is an ECC-origin skill that catalogs production-grade deployment workflows and CI/CD best practices for web applications. It activates when setting up CI/CD pipelines, Dockerizing services, choosing deployment strategies—rolling, blue-green, or canary—or implementing health checks, readiness probes, and environment-specific configuration before a production release. Developers reach for it to avoid reinventing rollback procedures, container layouts, and release checklists on every project. The skill packages opinionated patterns for production readiness rather than one-off shell snippets, making it a reference during infrastructure setup and release planning for SaaS and API backends.10.8kinstalls 37Continuous Learningcontinuous-learning is a legacy affaan-m/ecc skill marked DEPRECATED as of 2026-04-28 in favor of continuous-learning-v2. Version 1 used stop-hook observation to extract reusable patterns and instincts from Claude Code sessions into stored skills. continuous-learning-v2 is documented as a strict superset: stop-hook observation becomes PreToolUse and PostToolUse observation, full skills become atomic instincts with confidence scoring, and global-only storage becomes project-scoped with global promotion. Developers should route continuous learning, session learning, and pattern extraction requests to continuous-learning-v2 instead of invoking this v1 skill.10.7kinstalls 38Verification Loopverification-loop is a comprehensive verification skill from affaan-m/ecc for Claude Code sessions that enforces quality gates in six phases before merge-ready code ships. Phase 1 runs build verification via npm or pnpm build and stops on failure. Phase 2 type-checks TypeScript with npx tsc --noEmit or Python with pyright. Additional phases cover linting, tests, and further gates defined in the skill workflow. Invoke after completing a feature, before creating a PR, after refactoring, or whenever quality gates must pass systematically. Reach for verification-loop when ad hoc checks miss steps and you need a ordered stop-on-failure verification pipeline.10.7kinstalls 39Eval HarnessEval Harness is an ECC agent skill that formalizes eval-driven development (EDD) for Claude Code sessions. It lets developers define pass/fail criteria before work begins, measure agent reliability with pass@k metrics, and build regression test suites for prompt or agent changes. The skill activates when benchmarking agent performance across model versions or setting structured completion gates for AI-assisted workflows. Eval Harness treats evaluations as first-class artifacts rather than ad hoc manual checks. Developers reach for it when they need repeatable agent quality gates instead of one-off prompt tweaks.10.7kinstalls 40Deep Researchdeep-research is an ECC (Everything Claude Code) skill that produces thorough, cited research reports from multiple web sources. It activates Firecrawl and Exa MCP tools to search, crawl, and synthesize findings with source attribution. Use cases include competitive analysis, technology evaluation, market sizing, and due diligence on companies or investors. Developers reach for deep-research when a question needs synthesis across sources rather than a single documentation lookup. Trigger phrases include research, deep dive, investigate, and what's the landscape. Output is a structured report with citations agents and humans can verify.10.6kinstalls 41Cpp Testingcpp-testing is an affaan-m ECC agent skill for modern C++17 and C++20 test workflows using GoogleTest, GoogleMock, CMake, and CTest. It activates when developers write or fix C++ tests, design unit and integration coverage, configure CI gating, or diagnose failing and flaky cases. The skill also covers enabling AddressSanitizer and thread sanitizers for memory and race diagnostics during test runs. Developers reach for cpp-testing when C++ components need regression protection, consistent CMake test targets, or structured failure investigation rather than ad hoc compile-and-pray checks. It explicitly avoids non-test C++ feature work, keeping agents focused on test design, build integration, and reliability improvements ahead of release.10.6kinstalls 42Jpa PatternsJPA Patterns is an ECC-origin skill from affaan-m/ecc covering JPA and Hibernate work in Spring Boot applications. The skill activates when designing entities and table mappings, defining @OneToMany, @ManyToOne, and @ManyToMany relationships, optimizing queries with fetch strategies and projections to prevent N+1 problems, configuring transactions and auditing, implementing soft deletes, setting up pagination and custom repository methods, and tuning HikariCP connection pooling or second-level caching. Developers reach for JPA Patterns when scaffolding persistence layers or debugging slow Hibernate queries in production services. It consolidates entity design, relationship cardinality, indexing, and pooling guidance into one agent reference instead of scattered Stack Overflow threads.10.6kinstalls 43Search Firstsearch-first is a research-before-coding workflow skill from affaan-m/ecc that systematizes discovery of existing tools, libraries, MCP servers, and patterns before writing custom code. Developers reach for search-first when starting new features, adding dependencies or integrations, or when a user asks to add functionality and the agent is about to implement from scratch. The skill triggers on new feature work where existing solutions likely exist and guides agents to search repositories, package ecosystems, and MCP catalogs first. search-first prevents duplicate utilities, helpers, and abstractions by making discovery a mandatory gate before implementation.10.5kinstalls 44Clickhouse Ioclickhouse-io is an affaan-m/ecc agent skill covering ClickHouse database patterns, query optimization, and data engineering practices for analytical workloads. ClickHouse is a column-oriented OLAP DBMS optimized for fast analytics on large datasets with columnar storage, compression, parallel query execution, distributed queries, and real-time analysis. The skill documents MergeTree engine table design with Date partitioning, typed columns such as UInt64 volume and UInt32 trades, and markets_analytics-style schemas. Developers reach for clickhouse-io when designing analytical tables, writing performant aggregation queries, or tuning high-volume telemetry and market data pipelines. It focuses on OLAP-specific patterns rather than transactional CRUD database design.10.4kinstalls 45Configure Eccconfigure-ecc is an interactive step-by-step installer wizard for the Everything Claude Code (ECC) project. The skill uses AskUserQuestion to guide selective installation of skills and rules into user-level or project-level directories, validates paths and file correctness, and offers optimization passes on installed assets. Developers reach for configure-ecc when they say phrases like configure ecc, install ecc, or setup everything claude code, or when they want to verify or fix an existing ECC install. Bootstrap paths include installing via the ecc plugin with /plugin install ecc@ecc, which auto-loads the skill, or loading it directly from the ECC repository. The workflow covers fresh installs, selective skill picks, validation of broken paths, and project-specific tuning so agents load only what the workspace needs.10.4kinstalls 46Agentic Engineeringagentic-engineering is a skill from affaan-m/ecc for operating as an agentic engineer when AI agents handle most coding and humans enforce quality gates. Four operating principles require defining completion criteria before execution, decomposing work into agent-sized units, routing model tiers by task complexity, and measuring outcomes with evals and regression checks. The eval-first loop defines capability evals and regression suites before agents execute changes. Developers reach for agentic-engineering when shifting from manual implementation to supervised agent execution without sacrificing reliability, cost discipline, or verifiable completion criteria across multi-step software tasks.10.4kinstalls 47Kotlin Coroutines Flowskotlin-coroutines-flows is an affaan-m/ecc skill covering structured concurrency, Flow-based reactive streams, and coroutine testing for Android and Kotlin Multiplatform projects. It prescribes scope hierarchies from Application through viewModelScope down to structured child coroutineScope blocks, forbids GlobalScope, and documents patterns for parallel loads, debounce, retry, cancellation, and testing Flow emissions. Developers reach for kotlin-coroutines-flows when writing asynchronous Kotlin, wiring StateFlow or SharedFlow for reactive UI state, or debugging concurrency bugs in mobile and KMP codebases. The skill encodes production-ready coroutine idioms so agents generate cancellable, testable async code instead of fire-and-forget launches.10.4kinstalls 48Rust Patternsrust-patterns is an ECC-origin skill that encodes idiomatic Rust across six areas: ownership and borrowing, Result and `?` propagation with `thiserror` for libraries and `anyhow` for applications, traits, concurrency, and crate module layout. It activates when writing new Rust, reviewing pull requests, refactoring modules, or designing crate structure. The skill pushes compile-first output that follows community conventions instead of Java-style or overly verbose patterns. Developers reach for rust-patterns when they want agents to generate Rust that passes `cargo check` without iterative borrow-checker fixes.10.3kinstalls 49Iterative RetrievalIterative Retrieval is an ECC agent skill that solves the subagent context problem in multi-agent workflows where agents cannot know which files, patterns, or project terminology they need until work begins. Standard approaches fail by sending the entire codebase and exceeding token limits, sending nothing and leaving agents uninformed, or guessing required context and often missing critical files. Iterative Retrieval implements a four-phase loop—dispatch, evaluate, refine, and repeat—that progressively narrows retrieved context based on what each subagent discovers it still needs. Developers reach for Iterative Retrieval when delegating codebase tasks to subagents in large repositories where upfront context selection is unreliable and token budgets are tight.10.3kinstalls 50Laravel Tddlaravel-tdd is an affaan-m/ecc skill for test-driven development on Laravel applications using PHPUnit and Pest with an 80%+ coverage target across unit and feature tests. It follows a red-green-refactor cycle: write a failing test, implement code to pass, then refactor. The skill covers Eloquent models, policies, jobs, notifications, database factories, fakes, and new API endpoints. Pest is preferred for new tests unless the project is already standardized on PHPUnit. Developers reach for laravel-tdd when adding Laravel features, fixing bugs with regression tests, or refactoring with safety nets. It triggers on new endpoints, model changes, and any Laravel work where coverage and TDD discipline are required.10.3kinstalls 51Django Verificationdjango-verification is a verification-loop skill for Django projects from the affaan-m/ecc collection. The workflow walks through environment checks (Python version, virtualenv, DJANGO_SECRET_KEY), code quality with mypy, ruff, and Black, test execution with coverage, security scanning, and deployment readiness before merging or shipping. Developers reach for django-verification after large refactors, before opening a pull request, or ahead of a production deploy when manual checklists are easy to skip. The skill stops on misconfigured environments instead of masking failures, and encodes bash commands agents can run in sequence. It targets Django backends using pyproject.toml-driven tooling rather than greenfield scaffolding.10.2kinstalls 52Nutrient Document Processingnutrient-document-processing is an affaan-m/ecc skill for the Nutrient DWS Processor API at https://api.nutrient.io/build. Developers authenticate with a NUTRIENT_API_KEY and send multipart POST requests carrying an instructions JSON payload to convert formats, extract text and tables, OCR scanned documents, redact PII, add watermarks, apply digital signatures, and fill PDF forms. Supported input formats include PDF, DOCX, XLSX, PPTX, HTML, and images. Teams reach for nutrient-document-processing when building document pipelines—invoice ingestion, compliance redaction, or e-sign workflows—without maintaining local conversion libraries.10.2kinstalls 53Cost Aware Llm Pipelinecost-aware-llm-pipeline is an ECC agent skill that documents a composable Python pipeline for controlling LLM API spend while preserving quality. The pattern routes simple tasks to cheaper models such as claude-haiku-4-5-20251001 and reserves claude-sonnet-4-6 for workloads exceeding 10,000 characters or 30 items, with explicit thresholds in select_model(). Developers combine budget tracking, retry logic, and prompt caching when building Claude or GPT integrations, batching mixed-complexity items, or enforcing API budgets. Reach for cost-aware-llm-pipeline when LLM bills spike during development or when heterogeneous workloads need automatic tier selection instead of a single default model.10.2kinstalls 54Nextjs Turbopacknextjs-turbopack is a Claude Code skill from affaan-m/ecc covering Next.js 16+ local development with Turbopack, the Rust-based incremental bundler enabled by default for faster cold starts and hot module replacement. The skill explains when to stay on Turbopack for day-to-day work versus falling back to Webpack with --webpack or --no-turbopack when a Turbopack bug or webpack-only plugin blocks dev. It addresses incremental bundling, filesystem caching, and diagnosing slow HMR or startup in large apps. Developers reach for nextjs-turbopack when Next.js dev feels sluggish, when evaluating bundler tradeoffs, or when a legacy webpack plugin forces a temporary dev fallback while production builds may still use the default pipeline.10.1kinstalls 55Springboot Verificationspringboot-verification is an ECC-origin agent skill that chains Maven/Gradle build, static analysis, unit and integration tests with coverage thresholds, security scans, and diff review into one repeatable gate for Spring Boot microservices. Developers reach for it before opening a PR, after large refactors or dependency bumps, and immediately prior to staging or production deploys when a single missed regression or CVE would be costly. The skill encodes when to activate the full pipeline versus a lighter check and documents coverage thresholds and scan expectations so agents do not skip steps. It targets Java/Spring teams who want agent-driven CI parity on a laptop without wiring a custom script each sprint.10.1kinstalls 56Kotlin Patternskotlin-patterns is an ECC skill for idiomatic Kotlin development across seven key areas: null safety with the type system and safe-call operators, coroutines, DSL builders, module design, and Gradle Kotlin DSL configuration. Developers activate kotlin-patterns when writing new Kotlin, reviewing pull requests, refactoring legacy code, designing libraries, or tuning Gradle builds. The skill steers agents away from Java-style patterns toward Kotlin-first APIs, efficient concurrency, and type-safe null handling. It fits Android apps, Ktor or Spring Kotlin services, and shared multiplatform modules where consistent style reduces review churn. Use kotlin-patterns whenever Kotlin output must be production-grade rather than syntactically correct Java ports.10.1kinstalls 57Content Enginecontent-engine is an ECC agent skill that turns source material—articles, transcripts, demos, docs, and internal notes—into platform-native content systems that preserve the author's voice. The skill drafts X posts and threads, LinkedIn launch updates, short-form video scripts, YouTube explainers, newsletter editions, and repurposed multi-platform campaigns from a single asset. It activates when writing social posts, scripting video, building launch sequences, or adapting one source cleanly across X, LinkedIn, TikTok, YouTube, and email channels. content-engine emphasizes voice fidelity over flattened platform templates, structuring calendars and sequences for ongoing distribution rather than one-off copy dumps. Developers reach for content-engine after shipping features or recording demos when they need consistent public storytelling without rewriting every channel manually.10.1kinstalls 58Autonomous Loopsautonomous-loops is an ECC agent skill covering architectures for running Claude Code autonomously in loops. It spans simple sequential claude -p pipelines through RFC-driven multi-agent DAG systems with patterns and reference implementations teams can adopt. A v1.8.0 compatibility note states autonomous-loops is retained for one release while continuous-agent-loop becomes the canonical name for new loop guidance. Developers invoke autonomous-loops when choosing orchestration for long-running agent tasks, designing RFC-based multi-agent flows, or migrating existing ECC loop workflows without breaking references.10.1kinstalls 59Investor MaterialsInvestor Materials is an ECC agent skill for creating and updating investor-facing documents that stay internally consistent. It produces pitch decks, one-pagers, investor memos, accelerator applications, financial models, milestone plans, and use-of-funds tables tied to one shared source of truth. Developers reach for Investor Materials when fundraising assets drift out of sync or accelerator questionnaires need aligned projections. The skill emphasizes credible, defensible numbers across multiple artifacts rather than one-off slide copy. It activates during deck revisions, memo writing, and financial model updates.10kinstalls 60Swiftui Patternsswiftui-patterns is an ECC skill documenting modern SwiftUI architecture for Apple platforms using the Observation framework. It guides property-wrapper selection among @State, @Binding, and @Observable classes, designs NavigationStack navigation flows, structures view models and data flow, optimizes list and complex layout rendering, and applies environment values for dependency injection. Developers reach for swiftui-patterns when Claude or Cursor agents propose SwiftUI views that need consistent state boundaries, type-safe navigation, and performance-conscious composition instead of ad hoc UIKit-era patterns ported into declarative code.10kinstalls 61Ai First Engineeringai-first-engineering from affaan-m/ecc is an ECC skill for designing process, review, and architecture when AI-assisted code generation drives most implementation output. It shifts emphasis from typing speed to plan quality, from personal confidence to evaluation coverage, and from syntax nitpicks to system behavior during review. Architecture guidance favors explicit boundaries, stable contracts, typed interfaces, and deterministic tests over implicit conventions. Review targets regressions, security assumptions, data integrity, failure handling, and rollout safety while minimizing style debates already covered by automation. Strong adoption signals include decomposing ambiguous work, defining measurable acceptance criteria, writing high-signal prompts and evaluations, and enforcing risk controls under deadline pressure.9.9kinstalls 62Architecture Decision Recordsarchitecture-decision-records is an ECC-origin skill that turns coding-session architecture choices into structured ADR markdown following Michael Nygard's format adapted for AI-assisted development. It activates when developers explicitly ask to record a decision, select among frameworks, libraries, databases, or API patterns, or explain why X was chosen over Y. Each ADR includes date, status (proposed, accepted, deprecated), context, considered alternatives, and rationale so future contributors understand trade-offs without digging through Slack or PR threads. Use architecture-decision-records whenever a non-trivial structural choice should outlive the current Claude Code session.9.9kinstalls 63Bun Runtimebun-runtime is an ECC agent skill for Bun as a unified JavaScript toolkit: runtime, package manager, bundler, and test runner in one binary. It clarifies when to prefer Bun—new JS/TS projects, speed-sensitive scripts, Vercel deployments with Bun runtime, and single-toolchain workflows—and when to stay on Node for maximum ecosystem compatibility or legacy tooling dependencies. Developers reach for bun-runtime when adopting Bun, migrating from Node, debugging Bun-specific behavior, or configuring install, run, test, and build commands under one runtime.9.9kinstalls 64Context Budgetcontext-budget is an ECC skill that analyzes token overhead for every component loaded in a Claude Code session and surfaces practical optimizations to reclaim context space. Phase one inventories agents/*.md, skills/*/SKILL.md, MCP servers, and rules, estimating tokens per file as word count times 1.3 while flagging agent files over 200 lines and frontmatter descriptions over 30 words as bloated. Developers invoke it when session performance drops, output quality degrades after adding many skills or MCP servers, or when running the /context-budget command. The skill answers how much headroom remains before adding more components. context-budget suits maintainers curating large Claude Code workspaces who need data-driven trimming instead of guessing which SKILL.md files to disable.9.9kinstalls 65Regex Vs Llm Structured Textregex-vs-llm-structured-text is an affaan-m/ecc decision framework for parsing structured text such as quizzes, forms, invoices, and tables in agent and data pipeline contexts. The core insight is that regex handles 95-98% of cases deterministically at low cost, reserving expensive LLM calls for edge cases where pattern matching falls below a 95% success threshold. When over 90% of text follows repeatable patterns, developers start with regex and add LLM fallback only where confidence is low. Free-form or highly variable text goes directly to LLM. The skill covers hybrid architecture with a regex parser extracting structure first and an LLM layer handling exceptions. Developers reach for it when optimizing cost and reliability tradeoffs in text extraction pipelines.9.8kinstalls 66Content Hash Cache Patterncontent-hash-cache-pattern is an ECC skill for caching expensive file processing—PDF parsing, text extraction, image analysis—using SHA-256 content hashes as cache keys instead of file paths. Keys survive renames and moves and auto-invalidate when file bytes change. The pattern separates cache logic into a service layer and supports --cache/--no-cache CLI toggles. Developers reach for content-hash-cache-pattern when pipelines reprocess the same documents, processing cost is high, or agents add caching to file-ingestion tools. Unlike path-based caches, hash keys prevent stale hits after content edits even when filenames stay the same. Use it while building ingestion CLIs, document AI preprocessors, or batch analyzers.9.8kinstalls 67Continuous Agent Loopcontinuous-agent-loop is an ECC v1.8+ standard loop skill that replaces autonomous-loops while preserving compatibility for one release cycle. The skill routes work through a decision flow—continuous-pr for strict CI/PR control, rfc-dag for RFC decomposition, infinite for parallel exploration, or sequential as the default—and documents a production stack pairing RFC decomposition, plankton-code-quality gates, eval-harness evaluation, and nanoclaw-repl session persistence. Developers reach for continuous-agent-loop when agent runs churn without measurable progress, repeat the same root-cause retries, stall merge queues, or drift costs through unbounded escalation. The skill encodes failure modes and recovery patterns so agent loops terminate with evidence instead of infinite retries.9.8kinstalls 68Compose Multiplatform Patternscompose-multiplatform-patterns is an ECC skill cataloging production patterns for Jetpack Compose and Compose Multiplatform in Kotlin multiplatform projects. It covers state management with ViewModel plus single state data classes exposed as StateFlow, navigation setup, reusable composables, design-system theming, and recomposition performance tuning across Android, iOS, desktop, and Web targets. Developers reach for compose-multiplatform-patterns when implementing Compose UI, wiring ViewModel-driven screen state, building KMP navigation graphs, or optimizing render performance in shared UI codebases. Patterns emphasize a single screen-state data class collected in Compose rather than scattered mutable state. The skill activates whenever shared composable architecture decisions arise in KMP or Android-only Compose projects.9.8kinstalls 69Pytorch Patternspytorch-patterns is an Everything Claude Code reference for idiomatic PyTorch deep learning development. The skill encodes best practices for device-agnostic code that runs on CPU and GPU, robust training loops, efficient data loading, GPU memory optimization, and reproducible experiment configuration. Developers reach for pytorch-patterns when writing new models or training scripts, reviewing deep learning code, debugging training pipelines, or optimizing throughput and memory on accelerators. Core principles start with portable device handling and extend through structured patterns for model architecture, dataloaders, and experiment repeatability rather than ad-hoc notebook code.9.8kinstalls 70Flutter Dart Code Reviewflutter-dart-code-review is an ECC-origin agent skill that provides a comprehensive, library-agnostic checklist for reviewing Flutter and Dart applications before merge or ship. The checklist spans project health, widget best practices, state management patterns including BLoC, Riverpod, Provider, GetX, MobX, and Signals, Dart idioms, performance, accessibility, security, and clean architecture. It applies regardless of routing, DI, or state library choices and inspects pubspec.yaml hygiene, analysis_options.yaml lint strictness, and production code quality such as avoiding print() in favor of dart:developer logging. Developers reach for this skill when they need a consistent PR review rubric across Flutter codebases rather than ad-hoc comments.9.7kinstalls 71Ai Regression Testingai-regression-testing is an ECC agent skill for AI-assisted development that addresses the blind spot when the same model writes a fix and then approves its own review. The skill documents database-free sandbox API testing, automated bug-check workflows, and patterns for multi-path code such as sandbox versus production or feature-flag branches. Developers reach for ai-regression-testing after an AI agent edits API routes or backend logic, when a fixed bug needs a guard against recurrence, or when running commands like /bug-check after code changes. The workflow targets predictable failure modes where automated tests are the only systematic check that breaks shared assumptions between generation and review.9.7kinstalls 72Kotlin Testingkotlin-testing is a Claude Code skill from affaan-m/ecc that teaches and applies Kotlin testing patterns using Kotest, MockK, coroutine testing, property-based testing, and Kover coverage reporting while following strict TDD methodology. The workflow starts by identifying a target function, class, or module, then writes failing tests first, implements mocks and coroutine test dispatchers, and configures Kover for coverage thresholds. Developers reach for kotlin-testing when adding coverage to existing Kotlin services, writing new domain logic, or standardizing test style across a multi-module Gradle project. The skill emphasizes idiomatic Kotlin practices—structured concurrency tests, MockK verification, and maintainable spec layouts—rather than generic JUnit snippets ported from Java.9.7kinstalls 73Investor Outreachinvestor-outreach is a fundraising communication skill from affaan-m/ecc that generates short, concrete investor-facing messages instead of generic pitch spam. The skill activates when a developer needs a cold email to a partner, a warm intro request, a post-meeting follow-up, or a process update tailored to fund thesis and partner fit. Core rules enforce personalization on every outbound message and keep copy easy to act on with clear asks. investor-outreach fits technical founders who have product context but need high-signal wording for investor inboxes. Skip it when the task is product copy, hiring outreach, or customer sales emails rather than capital raising.9.7kinstalls 74Swift Actor Persistenceswift-actor-persistence is an ECC skill with patterns for thread-safe data persistence using Swift actors on Swift 5.9+ targeting iOS 17+ and macOS 14+. It combines in-memory caching with file-backed storage so shared mutable state stays isolated without manual locks or DispatchQueues. Developers activate it when building offline-first apps that need compile-time data-race safety instead of ad hoc synchronization. The actor model serializes access while preserving ergonomic Swift concurrency for read and write paths.9.7kinstalls 75Swift Protocol Di Testingswift-protocol-di-testing is an ECC-origin agent skill for protocol-based dependency injection in Swift codebases. The patterns abstract file system, network, external APIs, and iCloud access behind small focused protocols, enabling Swift Testing mocks that run deterministically across app, test, and SwiftUI preview targets. Reach for swift-protocol-di-testing when writing Swift that touches I/O, testing error paths without triggering real failures, or building modules that must behave identically in production, XCTest or Swift Testing, and preview environments. The skill emphasizes minimal protocol surface area so mocks stay maintainable as features grow.9.7kinstalls 76Documentation Lookupdocumentation-lookup is an ECC agent skill that routes library and framework questions to the Context7 MCP instead of relying on model training cutoffs. When a developer asks about React, Next.js, Prisma, or any named API, the skill calls Context7 tools resolve-library-id and query-docs to resolve a canonical library ID and return current documentation snippets and examples. The workflow replaces guesswork during setup, configuration, and API usage with source-aligned answers pulled at query time. Developers reach for documentation-lookup whenever they need version-accurate method signatures, migration notes, or official code samples while coding inside Claude Code, Cursor, or Codex. The skill activates on setup questions, API reference requests, and explicit framework names, making it a lightweight documentation bridge for full-stack and backend work without leaving the agent session.9.6kinstalls 77Blueprintblueprint is a construction-plan generator skill from affaan-m/ecc that converts a one-line objective into a step-by-step build plan any fresh coding agent can execute without prior context. The skill runs a 5-stage pipeline including research, dependency-graph ordering, parallel step detection, adversarial review gates, anti-pattern catalogs, and plan-change protocols. Developers reach for blueprint when splitting large features across multiple PRs, planning multi-session refactors or migrations, or coordinating parallel sub-agent work. blueprint explicitly avoids single-PR tasks completable in fewer than 3 tool calls or when the user says just do it.9.6kinstalls 78Design Systemdesign-system is a Design & UI/UX skill from affaan-m/ecc (origin: ECC) for building scalable, maintainable design systems with accessibility, responsive layouts, and theme configuration. Core elements include centralized design tokens for color, typography, spacing, and shadows; reusable button, input, and card components; layout patterns for pages, forms, and grids; and SVG icon libraries. The skill integrates Figma for design, Storybook for component showcase, Chromatic for visual testing, and JSON-managed design tokens. Its checklist covers token definition, base component implementation, Storybook setup, accessibility testing, documentation, and team adoption. Developers reach for design-system when initializing a component library or syncing design and implementation before release.9.6kinstalls 79Foundation Models On Devicefoundation-models-on-device in affaan-m/ecc documents patterns for integrating Apple's on-device language model on iOS 26 through the FoundationModels framework. Coverage spans text generation and summarization without cloud dependency, structured output via the @Generable macro, custom tool calling for domain-specific actions, and snapshot streaming for responsive UI updates—all running on-device for privacy and offline support. The skill addresses quantization, optimization, and privacy-conscious inference patterns suited to Apple Intelligence workloads in production Swift apps. Developers reach for foundation-models-on-device when shipping AI-powered iOS features that must avoid network round-trips or external API keys. Typical flows include extracting structured data from natural language, streaming partial model output into SwiftUI views, and wiring custom tools the on-device model can invoke locally. Triggers include FoundationModels setup, @Generable structured output, on-device LLM streaming, and Apple Intelligence integration without cloud APIs.9.6kinstalls 80Swift Concurrency 6 2swift-concurrency-6-2 is an ECC skill covering Swift 6.2 Accessible Concurrency, where code runs single-threaded by default and concurrency is introduced explicitly. It explains when to use @concurrent for background CPU work, how MainActor types maintain isolation consistency, and why Swift 6.1 implicit background offloading caused data-race errors in seemingly safe async code. Developers reach for swift-concurrency-6-2 when upgrading Swift 5.x or 6.0/6.1 projects to Swift 6.2, resolving Sendable and isolation diagnostics, or enabling Accessible Concurrency in Xcode 26 build settings. The skill targets performance-preserving patterns that eliminate common data races without over-offloading to background threads.9.6kinstalls 81Exa Searchexa-search is a Claude Code skill from affaan-m/ecc that routes research through the Exa MCP server's neural search engine. Developers activate it for current web information, code examples, API documentation references, company and competitor research, and professional people lookup during development tasks. Trigger phrases include search for, look up, find, and what's the latest, steering agents away from stale training data toward Exa-backed results. The skill fits background research for any build task where verified external sources matter more than model recall, and it documents when to prefer Exa over generic web search for technical and company queries.9.6kinstalls 82Backend PatternsBackend architecture patterns API design database optimization and server-side best practices for Node js Express and Next js API routes name backend-patterns description Backend architecture patterns API design database optimization and server-side best practices for Node js Express and Next js API routes Backend Development Patterns Backend architecture patterns and best practices for scalable server-side applications When to Activate Designing REST or GraphQL API endpoints Implementing repository service or controller layers Optimizing database queries N 1 indexing connection pooling Adding caching Redis in-memory HTTP cache headers Setting up background jobs or async processing Structuring error handling and validation for APIs Building middleware auth logging rate limiting API Design Patterns RESTful API Structure typescript PASS Resource-based URLs GET api markets List resources GET api markets id Get single resource POST api markets Create resource PUT api markets id Replace resource PATCH api markets id Update resource DELETE api markets id Delete resource PASS Query parameters for filtering sorting pagination GET api markets status active sort volume limit 20 offset 0 Rep.9.6kinstalls 83Jira Integrationjira-integration is an ECC skill that brings Jira issue management into AI coding workflows. It supports two paths: the recommended mcp-atlassian MCP server (Python 3.10+, uvx) and direct Jira REST API calls. Developers use jira-integration to fetch tickets for requirements, extract testable acceptance criteria, add progress comments, transition issues (To Do → In Progress → Done), link branches or merge requests, and search with JQL. The skill documents MCP configuration patterns and REST fallbacks so agents pick the right integration surface. Reach for jira-integration when implementation work is tracked in Jira and the agent must read ticket state or write back status during development.9.5kinstalls 84Data Scraper Agentdata-scraper-agent is a community ECC skill that builds production-ready, AI-powered data collection agents for public sources such as job boards, prices, news, GitHub repos, sports scores, and event listings. The generated stack uses Python, Gemini Flash for free LLM enrichment, GitHub Actions for scheduled runs at no hosting cost, and storage in Notion, Google Sheets, or Supabase. The agent scrapes on a schedule, enriches results with an LLM, persists output, and improves from user feedback over time. Developers reach for data-scraper-agent when they want to monitor or track public data without writing a scraper from scratch or paying for dedicated hosting infrastructure.9.5kinstalls 85Coding StandardsBaseline cross-project coding conventions for naming readability immutability and code-quality review Use detailed frontend or backend skills for framework-specific patterns name coding-standards description Baseline cross-project coding conventions for naming readability immutability and code-quality review Use detailed frontend or backend skills for framework-specific patterns Coding Standards Best Practices Baseline coding conventions applicable across projects This skill is the shared floor not the detailed framework playbook Use frontend-patterns for React state forms rendering and UI architecture Use backend-patterns or api-design for repository service layers endpoint design validation and server-specific concerns Use rules common coding-style md when you need the shortest reusable rule layer instead of a full skill walkthrough When to Activate Starting a new project or module Reviewing code for quality and maintainability Refactoring existing code to follow conventions Enforcing naming formatting or structural consistency Setting up linting formatting or type-checking rules Onboarding new contributors to coding conventions Scope Boundaries Activate this skill for descripti.9.5kinstalls 86Plankton Code Qualityplankton-code-quality is an ECC-origin skill that keeps code clean during active editing rather than deferring quality checks to pre-commit hooks or CI. It triggers formatting, lint passes, and Claude-assisted fixes immediately after file changes so developers see corrected code before pushing. Teams reach for plankton-code-quality when noisy lint debt accumulates between commits or when they want agent-driven fix suggestions inline with edits. The skill fits projects with established formatters and linters wired into the workspace. It complements but does not replace full test suites or security review pipelines.9.5kinstalls 87Visa Doc Translatevisa-doc-translate is an ECC agent skill that automates visa document localization from image inputs to submission-ready bilingual PDFs. When given an image path, the skill converts HEIC files to PNG with macOS sips, reads EXIF orientation and applies automatic rotation, then produces side-by-side original and English translation pages in a single PDF. The workflow runs without confirmation prompts so agents can batch embassy paperwork quickly. Developers reach for visa-doc-translate when supporting users who photograph foreign-language bank letters, employment certificates, or government forms and need consistent English translations packaged for consular review.9.5kinstalls 88Git Workflowgit-workflow in affaan-m/ecc documents Git collaboration patterns including GitHub Flow for continuous deployment, GitFlow for release branches, and trunk-based development options. The skill activates when setting up a new repository workflow, writing commit messages and PR descriptions, resolving merge conflicts, managing release tags, or onboarding contributors to shared Git practices. GitHub Flow is recommended for most teams with a protected main branch that stays deployable and short-lived feature branches merged via pull request. Examples illustrate branch naming, protected main policies, and PR description structure agents should follow when committing on behalf of developers. Developers reach for git-workflow when agent-assisted coding risks inconsistent branch hygiene or nonstandard commit messages across sessions. The skill originates from ECC and includes Japanese documentation in the source README while patterns apply to any Git-hosted project. Triggers include branching strategy choice, commit conventions, merge conflict resolution, and release tag management.9.5kinstalls 89Browser Qabrowser-qa is an agent skill from affaan-m/ecc (Everything Claude Code) that automates visual testing and UI interaction verification using browser automation MCP tools such as claude-in-chrome, Playwright, or Puppeteer. Phase 1 smoke tests navigate target URLs, filter console noise, check for 4xx/5xx network errors, capture desktop and mobile screenshots, and validate Core Web Vitals thresholds (LCP under 2.5s, CLS under 0.1, INP under 200ms). Phase 2 interaction tests click navigation links, submit forms with valid and invalid data, and verify error states. Developers reach for browser-qa after deploying to staging or preview, when reviewing frontend PRs, or before shipping to confirm layouts, forms, and interactions actually work in a real browser.9.5kinstalls 90Mcp Server Patternsmcp-server-patterns is an ECC skill for developers implementing or maintaining MCP servers using the Node/TypeScript SDK. The Model Context Protocol lets AI assistants call tools, read resources, and use prompts from a custom server, and this skill covers when to add each capability, how to choose stdio versus Streamable HTTP transport, and how to debug registration issues during SDK upgrades. Because the MCP SDK API evolves, the skill directs developers to Context7 MCP docs or official MCP documentation for current method names. Developers reach for mcp-server-patterns when wiring a new integration surface for coding agents—whether exposing an internal API, document store, or workflow engine—rather than hardcoding one-off scripts per agent session.9.5kinstalls 91Customs Trade Compliancecustoms-trade-compliance is an Apache-2.0 ECC skill at version 1.0.0 that role-plays a senior trade compliance specialist with 15+ years covering US, EU, UK, and Asia-Pacific customs operations. It codifies expertise for customs documentation, HS tariff classification, duty optimization, restricted-party screening, and multi-jurisdiction regulatory compliance inside AI agents that process import/export orders or supply-chain workflows. Referenced systems include ACE (Automated Commercial Environment), CHIEF/CDS (UK), ATLAS (DE), customs broker portals, screening platforms, and ERP trade management modules. Developers reach for customs-trade-compliance when agents must apply Incoterms, leverage FTAs, reduce penalties, and validate clearance logic—not generic shipping labels. It protects organizations from seizures and debarment while optimizing lawful cross-border movement. Use when handling customs clearance, tariff classification, trade compliance documents, or duty optimization in agent-driven commerce pipelines.9.4kinstalls 92Perl Securityperl-security is an ECC comprehensive security guide for Perl applications handling user input, filesystem paths, shell commands, and databases. Coverage includes taint mode (-T) for tracking untrusted external data, validation and untainting at input boundaries, constrained filesystem and process execution, parameterized DBI queries, and web protections against XSS, SQLi, and CSRF across CGI, Mojolicious, Dancer2, and Catalyst stacks. perlcritic security policies supplement manual review. Developers reach for perl-security when building or auditing Perl web apps and CLI tools that touch user input and must ship with hardened defaults instead of ad-hoc escaping.9.4kinstalls 93Fal Ai Mediafal-ai-media in affaan-m/ecc provides unified media generation through fal.ai MCP covering image, video, and audio workflows. Supported models include text-to-image Nano Banana, text and image-to-video with Seedance, Kling, and Veo 3, text-to-speech via CSM-1B, and video-to-audio with ThinkSound. Developers reach for fal-ai-media when users ask to generate images, create videos from text or images, produce speech or sound effects, or make thumbnails without leaving the agent environment. The skill activates on phrases like generate image, create video, text to speech, or make a thumbnail. MCP configuration is required to reach fal.ai endpoints from Claude Code or Cursor. Triggers span marketing assets, prototype media, and app placeholder content generated directly from prompts rather than stock libraries.9.4kinstalls 94Inventory Demand Planninginventory-demand-planning is an ECC skill (version 1.0.0, Apache-2.0) that codifies retail supply-chain expertise for demand forecasting, safety stock optimization, replenishment planning, and promotional lift estimation across multi-location retailers managing hundreds of SKUs. It guides forecasting method selection, ABC/XYZ segmentation, seasonal transition management, and vendor negotiation frameworks informed by long-running demand-planning practice. Developers and inventory-system builders reach for inventory-demand-planning when tuning ERP or WMS replenishment rules, modeling promotion spikes, or reducing stockouts and overstock without hiring a dedicated demand planner for every analysis cycle.9.4kinstalls 95Frontend PatternsGet battle-tested React, Next.js, and UI patterns that produce clean, performant, and maintainable frontend code.9.4kinstalls 96Enterprise Agent Opsenterprise-agent-ops is an ECC agent skill for operating cloud-hosted or continuously running agent systems beyond single CLI sessions. It structures four operational domains: runtime lifecycle with start, pause, stop, and restart controls; observability through logs, metrics, and traces; safety controls including scopes, permissions, and kill switches; and change management with rollout, rollback, and audit trails. Baseline controls include immutable deployment artifacts, least-privilege credentials, environment-level secret injection, hard timeout and retry budgets, and audit logs for high-risk actions. Developers reach for enterprise-agent-ops when production agents need operational guardrails, incident response hooks, and governed deployment practices rather than ephemeral local agent runs.9.4kinstalls 97Nestjs Patternsnestjs-patterns is a skill from affaan-m/ecc that codifies production-grade NestJS backend structure for modular TypeScript APIs. The skill activates when building NestJS services, organizing app.module.ts layouts with common filters and guards folders, adding DTO validation, configuring environment-aware settings, and testing HTTP endpoints. Developers reach for nestjs-patterns when they need opinionated folder conventions—src/main.ts, feature modules, and shared common utilities—instead of ad hoc NestJS scaffolding. It covers controllers, providers, guards, interceptors, exception filters, and database integrations as first-class building blocks for maintainable server code.9.4kinstalls 98Kotlin Ktor Patternskotlin-ktor-patterns is an ECC skill delivering comprehensive Ktor patterns for robust, maintainable HTTP servers built with Kotlin coroutines. It activates when building Ktor servers, configuring plugins (Auth, CORS, ContentNegotiation, StatusPages), implementing REST APIs, setting Koin dependency injection, writing testApplication integration tests, or adding WebSockets. The documented project layout places Application.kt at the entry point with plugins/Routing.kt, Serialization.kt, and related modules under src/main/kotlin. Developers reach for kotlin-ktor-patterns when agent-generated Ktor code drifts from idiomatic routing DSL, kotlinx.serialization setup, or coroutine-safe plugin configuration. It encodes standard structure and testing approaches so REST and WebSocket services stay consistent across modules. Use during Ktor greenfield APIs or refactors where Kotlin backend conventions must be preserved.9.4kinstalls 99Remotion Video Creationremotion-video-creation is a Remotion best-practices skill from affaan-m/ecc providing 29 domain-specific rules for creating animated videos in React. Rule files cover 3D content with Three.js and React Three Fiber, fundamental animations, assets, audio, captions, charts, transitions, and more. Developers reach for remotion-video-creation whenever agents touch Remotion code to avoid common pitfalls in composition timing, asset loading, and render pipelines. The skill supplies curated domain knowledge so coding agents produce production-quality Remotion projects rather than generic React patterns.9.4kinstalls 100Hexagonal Architecturehexagonal-architecture is a Ports and Adapters design skill that keeps domain logic independent from frameworks, transport layers, and persistence details. The core application depends on abstract ports, while adapters at the edges implement HTTP handlers, CLI commands, queue workers, cron jobs, databases, and external APIs. Use hexagonal-architecture when building features where long-term testability matters, refactoring layered code where I/O bleeds into domain logic, or supporting multiple interfaces for the same use case. The skill helps replace infrastructure such as databases or message brokers without rewriting business rules.9.4kinstalls 101Ralphinho Rfc Pipelineralphinho-rfc-pipeline is an ECC-origin skill inspired by humanplane-style RFC decomposition for multi-agent software delivery. Its seven pipeline stages cover RFC intake, DAG decomposition, unit assignment, unit implementation, unit validation, merge queue integration, and final system verification. Each work unit includes id, depends_on, scope, acceptance_tests, risk_level, and rollback_plan, with three complexity tiers from isolated file edits to multi-file behavior changes. Developers reach for ralphinho-rfc-pipeline when a feature is too large for one agent path and needs independently testable units executed safely in sequence across multiple agents.9.3kinstalls 102Quality Nonconformancequality-nonconformance is an ECC agent skill for building manufacturing quality workflows inside software: defect logging, root cause analysis, corrective and preventive actions, and audit-ready nonconformance records. The skill helps developers model investigation steps, escalation paths, and compliance traceability when shipping QMS, ERP, or MES features rather than ad-hoc ticket handling. Developers reach for quality-nonconformance when regulatory-grade quality processes must be encoded as structured application logic. The affaan-m/ecc readme is a Japanese translation placeholder referencing the English SKILL.md upstream. Use it when nonconformance management, CAPA, or defect RCA is a domain requirement in the codebase.9.3kinstalls 103Agent EvalAgent Eval is an ECC agent skill that systematizes head-to-head comparisons of coding agents on reproducible custom tasks. Tasks are declared in YAML specifying actions, target files, and success criteria, replacing subjective agent selection with measured pass rates, cost, runtime, and consistency metrics. Developers reach for Agent Eval before adopting a new tool or model, after agent updates, or when teams need data-backed agent choices on their own codebase. The skill supports Claude Code, Aider, Codex, and similar agents. Installation is performed from the ECC repository after reviewing source.9.3kinstalls 104Seoseo is an ECC-origin agent skill for technical SEO, on-page optimization, structured data, Core Web Vitals, and content strategy. It activates when developers audit crawlability and indexability, fix canonical and redirect issues, improve title tags and meta descriptions, add or validate schema markup, tune Core Web Vitals, run keyword research, map keywords to URLs, or plan internal links and sitemap or robots.txt changes. The skill prioritizes fixing technical blockers before content tweaks, enforces one primary search intent per page, and favors long-term quality signals over manipulative patterns with mobile-first assumptions. Recommendations are page-specific and implementation-ready for engineering handoff.9.3kinstalls 105Production Schedulingproduction-scheduling is an ECC agent skill that guides coding agents through manufacturing production scheduling, bottleneck resolution, and line-balancing decisions when building ERP, MES, or shop-floor tools. The skill translates operational scheduling expertise into implementable patterns for capacity planning, work-center assignment, and throughput optimization inside custom software. Developers reach for production-scheduling when modeling job queues, finite-capacity schedules, or constraint-based planners rather than generic CRUD backends. The published readme in affaan-m/ecc is a Japanese translation stub pointing to the English SKILL.md source, so triggers and workflow detail live in the upstream ECC skill file. Use it when manufacturing logic is a first-class domain concern in the codebase.9.3kinstalls 106Crosspostcrosspost is an ECC content-distribution skill that adapts a single underlying idea into platform-native posts for X, LinkedIn, Threads, and Bluesky using content-engine patterns. It never publishes identical copy across platforms, preserving voice while respecting each network's format and tone constraints. Developers reach for crosspost when a launch, release, update, or essay must appear on multiple social channels with tailored versions rather than one duplicated blast. Activation triggers include requests to crosspost, post everywhere, or adapt content for specific networks. The skill enforces a core rule: do not publish the same fake post in four costumes—each platform gets a distinct native rewrite of the same core message.9.3kinstalls 107Nanoclaw Replnanoclaw-repl is an ECC-origin skill for running persistent, branching REPL sessions that keep Claude agent work on-task across multiple turns and branches. It targets developers extending agent workflows who want REPL-like continuity instead of stateless prompt chains. Sessions remain focused on the active task while supporting branch exploration for alternatives or parallel hypotheses. The published readme excerpt is a Japanese translation stub pointing to the English SKILL.md source, so behavior is defined by the skill metadata and ECC agent patterns rather than the stub file alone.9.3kinstalls 108Videodbvideodb is an ECC agent skill that extends coding agents with end-to-end video intelligence: ingest from local files, URLs, RTSP feeds, or live desktop capture; build visual, semantic, and temporal indexes; search moments with timestamps and auto-clips; and act via transcoding, subtitle overlays, dubbing, translation, and real-time event alerts. The skill declares allowed tools Read, Grep, Glob, and Bash with python execution, signaling Python-driven media pipelines. Developers reach for videodb when building agents or apps that must process video, monitor live streams, or generate edited media assets without hand-rolling FFmpeg and search infrastructure. Use it for video RAG, stream monitoring, desktop session analysis, and programmatic timeline edits.9.3kinstalls 109Carrier Relationship Managementcarrier-relationship-management is an Apache-2.0 skill version 1.0.0 from affaan-m/ecc that encodes senior transportation manager expertise for carrier relationship workflows. It covers carrier portfolio management across truckload, LTL, intermodal, and brokerage modes for portfolios of 40–200+ active carriers, including sourcing, rate negotiation, RFP execution, routing guides, scorecard-based performance tracking, and contract renewals. The skill references TMS platforms, rate management tools, carrier onboarding portals, and market intelligence from DAT/Greenscreens. Developers building freight allocation, carrier scorecard, or rate strategy features reach for carrier-relationship-management to ground UI and business rules in operational logistics practice.9.3kinstalls 110Energy Procurementenergy-procurement is an Apache-2.0 licensed ECC skill at version 1.0.0 that packages codified expertise from energy procurement managers with 15+ years of experience at large commercial and industrial consumers. It guides tariff optimization, demand charge reduction, load profiling, hedging strategies, market structure analysis, renewable PPA evaluation, and sustainability reporting for multi-facility energy programs. Developers and platform teams running homelabs, colocation, or office/server facilities reach for energy-procurement when power contracts—not application code—drive a significant operating line item. The skill applies when procuring energy, comparing rate schedules, managing peak demand, or building facility-level cost strategies.9.3kinstalls 111Kotlin Exposed Patternskotlin-exposed-patterns in affaan-m/ecc documents comprehensive database access with JetBrains Exposed ORM for Kotlin backends. Exposed offers two query styles: a DSL resembling direct SQL for expressive queries and a DAO pattern for entity lifecycle management. HikariCP connection pooling is configured through HikariConfig for reusable database connections under load. Flyway runs versioned SQL migration scripts at startup to keep schemas synchronized across environments. The skill covers repository pattern implementations, JSON column handling, complex queries, and coroutine-safe transaction boundaries. Developers reach for kotlin-exposed-patterns when standing up Kotlin microservices, replacing JDBC boilerplate, or standardizing data access across a team. Typical deliverables include Table definitions, transaction wrappers, migration files, and repository interfaces wired to a pooled DataSource. Triggers include Exposed DSL setup, Flyway migrations in Kotlin, HikariCP configuration, and repository pattern with Exposed DAO.9.2kinstalls 112Logistics Exception Managementlogistics-exception-management is an Apache-2.0 ECC agent skill (version 1.0.0) that encodes 15+ years of freight operations expertise into structured exception-handling playbooks. The skill walks an agent through a five-step workflow: classify exceptions by type and severity, apply mode-specific resolution paths, document carrier evidence, escalate by dollar and time thresholds, and file claims inside statutory windows such as the 9-month Carmack Amendment deadline. Coverage spans six transport modes—LTL, FTL, parcel, intermodal, ocean, and air—with decision frameworks for eat-the-cost versus fight-the-claim calls, eight automatic escalation triggers, and eight weekly KPI targets. Developers reach for logistics-exception-management when building logistics SOP generators, claims automation, carrier-dispute assistants, or AI support for supply-chain operations teams integrating TMS, WMS, and ERP order systems.9.2kinstalls 113Brand Voicebrand-voice is an ECC-origin skill from affaan-m/ecc that constructs a durable voice profile from real source material—posts, essays, launch notes, documentation, or site copy—then reapplies that profile across content workflows. Instead of re-deriving tone each session or defaulting to generic AI phrasing, the skill captures diction, rhythm, and framing from authenticated examples. Developers and technical founders reach for it when writing for X, LinkedIn, email, launch posts, threads, or product updates in a recognizable voice. Activation triggers include requests for voice consistency, channel-specific outreach, or adapting a known author's tone. The output is a reusable style profile engineers can attach to downstream content, outreach, and social generation tasks inside agent sessions.9.2kinstalls 114Golang Testingname golang-testing description Go testing best practices including table-driven tests test helpers benchmarking race detection coverage analysis and integration testing patterns Use when writing or improving Go tests metadata origin ECC globs go go mod go sum Go Testing This skill provides comprehensive Go testing patterns extending common testing principles with Go-specific idioms Testing Framework Use the standard go test with table-driven tests as the primary pattern Table-Driven Tests The idiomatic Go testing pattern go func TestValidateEmail t testing T tests struct name string email string wantErr bool name valid email email user example com wantErr false name missing email userexample com wantErr true name empty string email wantErr true for _ tt range tests t Run tt name func t testing T err ValidateEmail tt email if err nil tt wantErr t Errorf ValidateEmail q error v wantErr v tt email err tt wantErr Benefits Easy to add new test cases Clear test case documentation Parallel test execution with t Parallel Isolated subtests with t Run Test Helpers Use9.2kinstalls 115Claude DevfleetClaude DevFleet is a multi-agent orchestration skill for Claude Code that connects to a running DevFleet MCP server at http://localhost:18801/mcp. Developers register the server with `claude mcp add devfleet --transport http http://localhost:18801/mcp`, then call plan_project to turn a prompt into a mission DAG, approve the plan, and dispatch missions so each agent works in its own git worktree with full tool access. Completed missions auto-merge before dependent missions dispatch, and get_report returns files_changed and structured completion data. Reach for Claude DevFleet when a single agent cannot safely parallelize backend, auth, tests, or refactors across branches.9.2kinstalls 116X Apix-api is an ECC skill for programmatic interaction with the X (Twitter) API v2 from Python agents and scripts. It documents OAuth 2.0 bearer tokens for read-heavy search and public data, plus OAuth 1.0a user context required for posting tweets, threads, DMs, and media uploads. Core operations include posting single tweets and reply chains, reading user timelines with public_metrics fields, recent search with query operators such as from:username and -is:retweet, username lookup, and v1.1 media upload followed by v2 tweet creation. Rate-limit handling reads x-rate-limit-remaining and x-rate-limit-reset headers at runtime rather than hardcoding static quotas, and the skill flags itself as drift-prone because X tiers and permissions change frequently. Security guidance forbids hardcoded tokens, requires .gitignore for .env files, and recommends read-only credentials when writes are unnecessary. Developers reach for x-api when agents or scripts must post threads, search recent tweets, or pull engagement metrics from X.9.2kinstalls 117Dmux Workflowsdmux-workflows is an affaan-m/ecc skill for multi-agent orchestration using dmux, a tmux pane manager built for AI agent harnesses. It defines patterns to run Claude Code, Codex, OpenCode, and other agents in parallel panes, split large tasks, and merge results from concurrent sessions. Activate when a user says run in parallel, split this work, use dmux, or coordinate multi-agent development on a complex refactor or feature slice. The skill covers pane layout, work partitioning, and cross-harness coordination rather than single-threaded prompting. Developers with tmux-based terminals and multiple agent CLIs gain the most from these workflows.9.2kinstalls 118Team Builderteam-builder is an affaan-m/ecc agent skill that presents an interactive menu to browse a collection of agent persona markdown files and assemble parallel teams for a task. Agent files must include role, identity, rules, workflow, and deliverable sections; the first heading becomes the agent name and the first paragraph becomes its description. The skill supports flat layouts and domain subdirectories where folder names infer specialties such as engineering, marketing, or sales. Developers reach for team-builder when they have many agent roles and need to mix domains—security plus SEO plus architecture—before committing to a workflow. It does not run agents itself; it helps choose and compose the right roster from existing markdown definitions.9.2kinstalls 119Returns Reverse Logisticsreturns-reverse-logistics is an ECC skill (version 1.0.0, Apache-2.0) encoding a senior returns operations manager playbook for retail, ecommerce, and omnichannel environments. It covers RMA approval, receiving and inspection, condition grading, disposition routing, refunds and credits, fraud detection, vendor return-to-vendor recovery, and warranty claims across OMS, WMS, RMS, CRM, and fraud platforms. The framework balances customer satisfaction with margin protection and inspection accuracy. Developers reach for it when designing return policies, automation rules, or support playbooks rather than hiring dedicated reverse-logistics staff.9.2kinstalls 120Perl Patternsperl-patterns is an ECC skill documenting idiomatic Modern Perl 5.36+ patterns for robust, maintainable applications. A single use v5.36 pragma replaces older boilerplate by enabling strict, warnings, and subroutine signatures. The guide biases toward explicit modules, centralized error handling, and testable boundaries when writing new code, reviewing for idiomatic compliance, or migrating pre-5.36 legacy Perl. Developers reach for perl-patterns when authoring Perl modules, designing architecture, or refactoring legacy scripts that still use antiquated pragmas instead of signature-based Modern Perl defaults.9.2kinstalls 121Nuxt4 Patternsnuxt4-patterns is an ECC skill for building and debugging Nuxt 4 applications that mix SSR, hybrid rendering, and client-only islands. Hydration guidance keeps the first render deterministic by moving Date.now(), Math.random(), browser APIs, and storage reads behind onMounted, import.meta.client, ClientOnly, or .client.vue components. Data-fetching rules prefer await useFetch for SSR-safe reads, useAsyncData with stable keys for composed sources, and reserve $fetch for user-triggered writes. Route rules in nuxt.config.ts cover prerender, swr, isr, ssr false, and Nitro cache strategies per route group rather than globally. Performance patterns use Lazy-prefixed dynamic imports, hydrate-on-visible lazy hydration, and NuxtLink prefetching. A five-item review checklist verifies matching server and client markup, correct fetch composables, lazy non-critical data, aligned route rules, and lazy-loaded interactive islands. Developers reach for nuxt4-patterns when hydration warnings, route-rule tuning, or payload bloat appear in Nuxt 4 apps.9.2kinstalls 122Perl Testingperl-testing is an ECC-origin agent skill for writing reliable Perl modules and applications through strict test-driven development. The workflow walks developers through Test2::V0 and Test::More harness setup, red-green-refactor cycles, and assertion patterns so CPAN-style modules and CLI tools ship with deterministic coverage. Reach for perl-testing when adding Perl features, fixing regressions, or refactoring legacy scripts where missing tests block safe changes. The skill assumes familiarity with Perl's testing ecosystem and focuses on module boundaries, mockable dependencies, and CI-friendly test layout rather than application scaffolding from scratch.9.1kinstalls 123Laravel Verificationlaravel-verification is an ECC skill that runs a complete Laravel quality gate before pull requests, major refactors, or production deploys. The loop spans eight sequential phases: environment and Artisan checks, Composer validation, Pint linting, PHPStan or Psalm static analysis, PHPUnit tests with optional Xdebug coverage, composer audit security scans, migration pretend runs with rollback review, and deployment readiness including config, route, and view cache warmups plus queue and scheduler checks. Each phase gates the next—failed environment or Composer checks stop the pipeline before expensive test runs. The skill documents Sail-compatible commands, Horizon status checks, queue:monitor backlog inspection, and a staging-only queue healthcheck job pattern. CI examples chain pint --test, phpstan analyse, and coverage-enabled php artisan test. Developers reach for laravel-verification when Laravel changes need proof that lint, tests, security, migrations, and release caches all pass together.9kinstalls 124Rules Distillrules-distill is an affaan-m/ecc meta-skill that scans installed skills, extracts cross-cutting principles appearing in multiple skills, and distills them into agent rule files by appending, revising outdated content, or creating new rules. It applies a deterministic collection plus LLM judgment pattern: scripts collect facts exhaustively, then an LLM cross-reads full context and produces verdicts. Developers reach for rules-distill during periodic rules maintenance, monthly upkeep, or after installing new skills when a skill stocktake reveals overlapping guidance. The skill prevents rule drift and duplication across a growing skill library. It is designed for teams accumulating many agent skills who need consolidated, maintainable rule governance.9kinstalls 125Skill Complyskill-comply is an automated compliance measurement skill for coding agents. It auto-generates expected behavioral sequences from any .md skill, rule, or agent definition, then builds scenarios at three prompt strictness levels: supportive, neutral, and competing. The skill runs `claude -p`, captures tool call traces via stream-js, classifies behavioral sequences, and reports compliance rates with full tool call timelines. Use skill-comply when custom agent instructions exist but behavior consistency is unverified. Required tools are Read and Bash, making it suited to CI-style agent definition audits before wider rollout.8.9kinstalls 126Accessibilityaccessibility is an ECC agent skill focused on WCAG 2.2 Level AA technical implementation across web, iOS, and Android. It applies POUR principles—Perceivable, Operable, Understandable, and Robust—to translate high-level design requirements into concrete attributes such as ARIA roles, accessibility traits, hints, and focus behavior. Developers reach for accessibility when defining new component specs, implementing WCAG 2.2 criteria like Target Size and Focus Appearance, or auditing existing code for screen reader, keyboard, and switch-control barriers. The skill emphasizes semantic mapping over generic containers so assistive technologies read a correct accessibility tree. Use it during component design, pre-release audits, and remediation sprints where compliance gaps must become actionable implementation tasks.8.9kinstalls 127Click Path Auditclick-path-audit is a behavioral flow audit skill from affaan-m/ecc that finds bugs static reading and unit tests miss. Traditional checks verify functions exist, avoid crashes, and return correct types, but not whether final UI state matches button labels or whether one handler silently undoes another. The skill traces user-facing touchpoints through ordered calls, race conditions, and shared store side effects in Zustand, Redux, or context. Developers reach for click-path-audit when users report broken buttons despite green tests, especially after refactors touching global state. Example failure modes include compose mode toggles canceled by thread selection side effects.8.9kinstalls 128Santa Methodsanta-method is an affaan-m/ecc adversarial verification framework originating from Ronald Skelton's RapportScore.ai work. A single self-reviewing agent shares the same biases and knowledge gaps as the generator; Santa Method breaks that failure mode by spawning two independent reviewers without shared context who must both approve before output ships. Developers invoke santa-method before publishing customer-facing copy, deploying code without human review, or releasing compliance-sensitive, legally worded, or hallucination-prone content such as API references and statistics. The skill iterates fixes until both reviewers pass, but is not intended for internal drafts, exploratory research, or tasks already covered by build, test, and lint pipelines.8.8kinstalls 129Councilcouncil is an ECC agent skill that gathers four perspectives—contextual Claude, a skeptic sub-agent, a pragmatist sub-agent, and a critic sub-agent—to challenge conversational anchoring on fuzzy decisions. It is designed for moments with multiple credible paths and no obvious winner, such as monorepo versus polyrepo, ship now versus delay for polish, feature flags versus full rollout, or simplifying scope versus keeping strategic breadth. council explicitly excludes output verification, feature implementation planning, and system architecture design, pointing those tasks to other ECC skills. Developers reach for council when they need explicit tradeoff visibility, opposing viewpoints, or adversarial challenge before a go/no-go call.8.8kinstalls 130Token Budget Advisortoken-budget-advisor, abbreviated TBA, intercepts the response flow before Claude answers so users pick how much depth to consume. Trigger phrases include token budget, token count, response length, answer depth, short version, detailed answer, exhaustive answer, and explicit percentage requests such as respond at 50%. The skill does not fire when the user already set a level for the session, wants a one-word reply, or refers to auth or payment tokens instead of LLM tokens. Use token-budget-advisor when answer size should be negotiated up front to save tokens on large explanations or reserve depth for complex tasks.8.8kinstalls 131Dashboard Builderdashboard-builder is version 1.0.0 ECC skill adapted for Grafana, SigNoz, and similar observability platforms. It starts from operator questions—Is it healthy? Where is the bottleneck? What changed? What action should someone take?—rather than dumping every available metric into unstructured panels. Guardrails forbid mixing health, throughput, and resource panels without structure, and require titles, units, and reasonable thresholds on every panel. Developers reach for dashboard-builder when asked to build Kafka monitoring dashboards, create Grafana boards for Elasticsearch, stand up SigNoz dashboards for a service, or turn a raw metrics list into a functioning operational dashboard operators can act on.8.8kinstalls 132Ui Demoui-demo is a Playwright-based demo recording skill from affaan-m/ecc that produces polished WebM walkthrough videos of web applications with injected visible cursor overlays, natural pacing, and narrative flow. The skill follows a mandatory 3-phase process—explore, rehearse, record—never jumping straight to recording without first mapping real page elements. Developers reach for ui-demo when users request demo videos, screen recordings, walkthroughs, or tutorial captures for documentation, onboarding, or stakeholder presentations. ui-demo inspects actual DOM structure before scripting to avoid assumptions about inputs, dropdowns, and interactive components.8.8kinstalls 133Dart Flutter Patternsdart-flutter-patterns is an ECC skill with production-ready, copy-paste Dart and Flutter patterns organized by concern. Coverage includes null safety (avoiding `!`, preferring `?.`/`??`/pattern matching), immutable state with sealed types, freezed, and copyWith, async composition, widget architecture, BLoC, Riverpod, and Provider state management, GoRouter navigation with auth guards, Dio HTTP clients, Freezed code generation, WebView integration, local storage, and widget/Cubit/Riverpod tests. Developers reach for dart-flutter-patterns when starting Flutter features, choosing state libraries, wiring authenticated routes, or reviewing Dart for null-safety and sealed-type conventions. The skill accelerates consistent mobile code instead of mixing ad hoc patterns across modules.8.8kinstalls 134Healthcare Phi Compliancehealthcare-phi-compliance is an affaan-m/ecc skill (version 1.0.0) contributed from healthcare practice, providing patterns to protect patient, clinician, and financial data in clinical systems. It guides access control, authentication, database schema design, API response filtering, audit trails, and logging when building features that touch patient records. The skill spans HIPAA (US), DISHA (India), and GDPR (EU) requirements alongside general healthcare data protection. Developers invoke it when designing EHR integrations, patient portals, or clinician dashboards where PHI leakage, weak authorization, or non-compliant logs create regulatory risk. Outputs include concrete schema, API, and logging patterns rather than generic security checklists.8.8kinstalls 135BenchmarkBenchmark is an ECC performance skill for measuring baselines, catching PR regressions, and comparing stack alternatives. Mode 1 drives a real browser via browser MCP to capture Core Web Vitals—LCP under 2.5s, CLS under 0.1, INP under 200ms, FCP under 1.8s, and TTFB under 800ms—plus total page weight under 1MB and JS bundles under 200KB gzipped. Developers reach for Benchmark before launch, when users report slowness, or when evaluating framework swaps. The skill turns subjective performance complaints into measured thresholds suitable for CI-style gates.8.8kinstalls 136Manim Videomanim-video is an ECC-origin Claude Code skill for building reusable Manim explainers where motion, structure, and clarity matter more than photorealism. It guides agents to scaffold scenes for graphs, workflows, architecture diagrams, metric progressions, and short product or launch explainers suitable for X posts or landing pages. The skill integrates with the wider ECC video stack when a project needs compositing or handoff beyond raw Manim output. Developers reach for manim-video when a concept is inherently spatial or temporal and static screenshots fail to communicate behavior, topology, or step-by-step flow.8.7kinstalls 137Healthcare Emr Patternshealthcare-emr-patterns is version 1.0.0 from affaan-m/ecc, contributed by Health1 Super Speciality Hospitals via Dr. Keyur Patel, for building Electronic Medical Record and Electronic Health Record systems. It prioritizes patient safety, clinical accuracy, and practitioner efficiency across encounter workflows—from complaint and exam through diagnosis and prescription—and clinical note-taking with structured fields, free text, and voice-to-text. The skill covers prescription modules with drug interaction checking, Clinical Decision Support Systems, and HL7/FHIR integration patterns. Developers reach for healthcare-emr-patterns when implementing regulated clinical UI flows rather than generic form builders.8.7kinstalls 138Safety Guardsafety-guard is an ECC protection skill for autonomous agent runs on production systems and sensitive repositories. It provides three protection modes: Careful mode detects destructive commands such as rm -rf, git push --force, DROP TABLE, kubectl delete, and --no-verify hooks before execution and prompts for confirmation with safer alternatives; Freeze mode locks edits to specific directory trees via commands like freeze src. Developers reach for safety-guard during full-auto agent sessions, migrations, deploys, and data changes where a mistaken shell or Git command could wipe state. The skill watches roughly a dozen high-risk command patterns and offers directory-scoped edit locks instead of relying on post-hoc review.8.7kinstalls 139Repo Scanrepo-scan is a cross-stack source audit skill that classifies each file, detects third-party libraries embedded without package managers, and assigns practical four-level verdicts per module alongside an interactive HTML report. Developers install it from haibindev/repo-scan via a pinned git fetch into ~/.claude/skills/repo-scan for reproducible runs. Use repo-scan when taking over large legacy codebases, planning refactors to separate core from duplicate or dead code, auditing vendored C++, Android, iOS, or Web dependencies, or preparing architecture decision records for monorepo reorganization. The skill addresses ecosystems where no single dependency manager explains what code is owned versus third-party.8.7kinstalls 140Laravel Plugin Discoverylaravel-plugin-discovery is an ECC skill for finding, evaluating, and choosing healthy Laravel packages using the LaraPlugins.io MCP server. Developers use it when searching packages for features like auth, permissions, or admin panels; asking what package to use for a capability; checking active maintenance; verifying Laravel version compatibility; or assessing package health before composer require. The LaraPlugins MCP server must be configured in the Claude environment—typically under ~/.claude MCP settings. The skill covers package discovery, dependency resolution context, and service provider integration planning without blindly adding unmaintained forks. Reach for it early in Laravel feature spikes when multiple Packagist options exist and compatibility with the project's Laravel version must be confirmed before integration work begins.8.7kinstalls 141Product Lensproduct-lens is an ECC-origin skill designed to challenge product assumptions and surface the underlying why before agents translate requests into code. Developers reach for product-lens when stakeholders hand over underspecified features, when the motivation behind a ticket is unclear, or when a quick sanity check could prevent building the wrong thing. The skill acts as a pre-build lens on intent, forcing clarity on user problems, success criteria, and tradeoffs instead of jumping straight to files and functions. It fits early validation workflows alongside prototyping and scoping skills. The published readme notes an in-progress Japanese translation, but the skill definition and description remain active in the ECC catalog for English-language product questioning sessions inside Claude Code agent workflows.8.6kinstalls 142Agent Payment X402agent-payment-x402 is an agent payments skill from affaan-m/ecc that adds policy-gated spending through the x402 HTTP payment protocol and MCP tools. The skill supports non-custodial wallets via agentwallet-sdk on Base and OKX Payments on X Layer without custodial risk. Use cases include agents paying for API calls, purchasing services, settling with other agents, and enforcing per-task spend limits. A decision table routes integrations depending on whether the agent buys access to 402-gated APIs or charges others. agent-payment-x402 pairs naturally with cost-aware-llm-pipeline and security-review skills during paid agent workflows.8.6kinstalls 143Ckck (Context Keeper) is a version 2.0.0 community agent skill by sreedhargs89 that gives Claude Code persistent per-project memory loading automatically at every session start. Scripts live at ~/.claude/skills/ck/commands/ and run deterministically via /ck:* commands, presenting stdout verbatim for consistent behavior across model versions. The skill tracks sessions with git activity and writes to native Claude memory under ~/.claude/. Developers reach for ck when context resets between sessions break continuity on long-running repositories. Invoke any /ck:* command to run the matching Node.js script rather than relying on model recall alone.8.6kinstalls 144Agent Introspection DebuggingAgent Introspection Debugging is an ECC workflow skill for structured self-debugging when AI agent runs stall. It activates on maximum tool-call or loop-limit failures, repeated retries without progress, context growth, prompt drift, and filesystem errors. The workflow captures failure state, diagnoses root causes, attempts contained recovery, and produces introspection reports before escalating to a human. Developers reach for Agent Introspection Debugging when agents consume tokens without forward motion or revisit the same tools endlessly. The skill is an explicit workflow, not a hidden runtime patch.8.6kinstalls 145Canary WatchCanary Watch is an ECC agent skill that monitors a deployed URL for regressions after production or staging deploys, risky PR merges, dependency upgrades, or during launch windows. It runs in a loop until stopped or until the monitoring window expires, checking six signal categories: HTTP status codes, new browser console errors, failed network and 5xx API calls, LCP/CLS/INP performance against baseline, missing critical DOM elements such as h1, nav, footer, and CTA, and API endpoint health within SLA. Quick-check mode runs a single pass and reports results via `/canary-watch https://myapp.com`, while continuous mode rechecks every N minutes for M hours with `--interval 5m`. Developers reach for Canary Watch when they need automated post-deploy confidence without standing up a full APM stack.8.6kinstalls 146Nodejs Keccak256nodejs-keccak256 is an ECC skill at version 1.0.0 that stops a silent Ethereum hashing bug in Node.js and TypeScript codebases. Node's crypto.createHash('sha3-256') implements the NIST-standardized SHA3 variant, not Ethereum Keccak-256, and produces different digests for the same input without warnings—breaking function selectors, event topics, EIP-712 signatures, storage slots, and address derivation. The skill documents correct patterns with ethers v6 keccak256 and id helpers, viem toBytes hashing, and web3.js utils.keccak256, plus mapping slot encoding and public-key-to-address derivation. It includes grep audit commands to find createHash('sha3-256') misuse across .ts and .js files and worked examples for solidityPackedKeccak256 and Transfer event topics. Developers reach for nodejs-keccak256 when reviewing Web3 backends or building Merkle, signature, or ABI selector utilities in JavaScript.8.5kinstalls 147Healthcare Cdss Patternshealthcare-cdss-patterns is an ECC agent skill at version 1.0.0 contributed from Health1 Super Speciality Hospitals for building Clinical Decision Support Systems integrated into EMR workflows. CDSS modules are patient-safety critical with zero tolerance for false negatives. The skill covers drug interaction checking, dose validation engines, clinical scoring systems including NEWS2, qSOFA, APACHE, and GCS, alert systems for abnormal clinical values, medication order entry safety checks, and lab result interpretation hooks. Developers reach for healthcare-cdss-patterns when implementing evidence-based clinical logic in hospital or health-tech backends rather than generic CRUD APIs. Use it when alerts, scoring, and medication safety checks must meet clinical reliability standards.8.5kinstalls 148Dotnet Patternsdotnet-patterns is an ECC-origin agent skill covering idiomatic C# and .NET conventions for building robust, performant, maintainable applications with ASP.NET Core. Developers activate it when writing new C# code, reviewing pull requests, refactoring legacy .NET apps, or designing service architectures with proper dependency injection boundaries. Core guidance includes preferring immutable records and init-only properties, structured async/await usage, sealed record value objects like Money DTOs, and clean separation in service layers. The skill encodes when-to-activate triggers for new code, reviews, and refactors so agents apply consistent .NET idioms instead of generic OOP patterns from other languages.8.5kinstalls 149Security Bounty Huntersecurity-bounty-hunter is an ECC skill (version 1.0.0) that focuses security review on exploitable, remotely reachable attack paths platforms like HackerOne and Huntr actually pay for. The workflow deprioritizes noisy local-only issues and patterns bounty platforms routinely reject as informational or out of scope. Documented in-scope patterns include user-controlled URL SSRF (CWE-918), middleware or API guard authentication bypass (CWE-287), and remote deserialization chains with concrete CWE mappings and impact notes. Developers reach for security-bounty-hunter when triaging a repo for paid bounty eligibility rather than running a generic secure-coding checklist.8.5kinstalls 150Healthcare Eval Harnesshealthcare-eval-harness is a version 1.0.0 ECC skill contributed from Health1 Super Speciality Hospitals that implements an automated verification system for healthcare application deployments. A single CRITICAL failure blocks deployment because patient safety is non-negotiable. It validates healthcare AI models, clinical metrics, and regulatory compliance checks before EMR/EHR releases or after CDSS logic changes such as drug-interaction rules. Examples reference Jest as the test runner, but categories and pass thresholds are framework-agnostic and adapt to Vitest, pytest, or PHPUnit. Developers reach for healthcare-eval-harness when agents touch clinical decision support, protected health information, or regulated hospital software where manual QA cannot catch safety regressions. The harness turns patient safety, CDSS accuracy, and PHI protection into gateable automated tests instead of checklist reminders.8.5kinstalls 151Customer Billing Opscustomer-billing-ops is an ECC agent skill for real customer billing operations, not generic payment API design. It walks developers and support engineers through Stripe-first workflows to answer who the customer is, what failed, the safest fix, and what follow-up to send. The skill covers refund handling, churn triage, duplicate or accidental charges, failed renewals, subscription and invoice inspection, plan-mix review, and billing-portal flow validation. It prioritizes connected billing tools like Stripe and hosted customer portals over custom account-management code, and treats email, GitHub, or issue trackers as supplemental evidence only. Reach for customer-billing-ops when a paying user reports broken billing, cancellation problems, or subscription confusion that directly affects revenue.8.5kinstalls 152Github Opsgithub-ops is a skill from affaan-m/ecc for managing GitHub repositories with a focus on community health, CI reliability, and contributor experience. The skill activates when triaging and labeling issues, checking PR review and CI status, debugging broken pipelines, preparing releases and changelogs, or monitoring Dependabot and security alerts. All GitHub API operations go through the gh CLI as the required tool. Developers reach for github-ops when they say check GitHub, triage issues, review PRs, merge, release, or CI is broken and need structured automation instead of ad hoc web UI clicks across mature repositories.8.4kinstalls 153Lead Intelligencelead-intelligence is an ECC agent-powered pipeline for finding, qualifying, and reaching high-value contacts through social graph analysis and warm-path discovery. It replaces manual prospecting workflows with signal scoring, mutual ranking, source-derived voice modeling, and outreach tailored to email, LinkedIn, and X channels. Developers or technical founders reach for lead-intelligence when building partnership lists, investor outreach, or industry-specific prospect pipelines without traditional enrichment SaaS. Activation covers requests to find leads, qualify prospects, or launch outreach in a target industry. The skill models authentic voice from source material and prioritizes warm introduction paths over cold blasts.8.4kinstalls 154Springboot PatternsSpring Boot architecture patterns REST API design layered services data access caching async processing and logging Use for Java Spring Boot backend work name springboot-patterns description Spring Boot architecture patterns REST API design layered services data access caching async processing and logging Use for Java Spring Boot backend work origin ECC Spring Boot Development Patterns Spring Boot architecture and API patterns for scalable production-grade services When to Activate Building REST APIs with Spring MVC or WebFlux Structuring controller service repository layers Configuring Spring Data JPA caching or async processing Adding validation exception handling or pagination Setting up profiles for dev staging production environments Implementing event-driven patterns with Spring Events or Kafka REST API Structure java RestController RequestMapping api markets Validated class MarketController private final MarketService marketService MarketController MarketService marketService this marketService marketService GetMapping ResponseEntity Page MarketResponse list RequestParam defaultValue 0 int page RequestParam defaultValue 20 int size Page Market markets marketService list Pag.8.4kinstalls 155Finance Billing Opsfinance-billing-ops is an ECC workflow skill for operators and backend developers who need code-backed truth about how money moves through a product. Instead of generic payment advice, it walks through revenue state, pricing decisions, team seat logic, duplicate-charge diagnosis, and refund behavior by pulling evidence from implementation details, sales snapshots, and related repositories. The skill composes with five related ECC skills—customer-billing-ops, research-ops, market-research, github-ops, and verification-loop—when competitive pricing, backlog state, or checkout proof is required. Reach for finance-billing-ops when website copy, dashboards, or support tickets disagree with what billing code enforces.8.4kinstalls 156Api Connector Builderapi-connector-builder is an ECC agent skill (version 1.0.0) for adding repository-native API connectors without inventing a parallel integration architecture. Agents first study in-repo connectors—layout, configuration schema, authentication, error handling, test style, and registration wiring—then implement new providers such as Jira or Slack to match those conventions. The skill blocks generic HTTP-client shortcuts and vendor-doc-only starts when the repo already defines integration patterns. Developers use api-connector-builder when extending multi-connector codebases that expect full transport, registry, test, and documentation coverage per integration.8.4kinstalls 157Openclaw Persona Forgeopenclaw-persona-forge is an ECC community skill that forges full OpenClaw lobster-agent personas from guided Q&A or random gacha mode. It outputs identity positioning, a SOUL.md soul description, character boundary rules, a chosen name, and an image-generation prompt for a unified-style avatar; when an approved image skill is available, it can produce the avatar file too. Developers reach for openclaw-persona-forge when bootstrapping a new OpenClaw agent that needs personality, not just tool wiring. The skill explicitly avoids fine-tuning existing SOUL.md files, non-OpenClaw frameworks, and purely utilitarian agents without character. Trigger phrases include lobster soul, OpenClaw SOUL, and gacha persona generation.8.4kinstalls 158Google Workspace Opsgoogle-workspace-ops is an ECC skill for treating Google Workspace as an integrated workflow system rather than editing isolated files. It uses Google Drive as the entry point, then routes work to Docs, Sheets, or Slides for in-place updates, consolidation, restructuring, import repair, and decision-ready summaries. Developers reach for google-workspace-ops when locating shared plans or customer lists in Drive, cleaning messy spreadsheets, reformatting slide decks, or producing summaries from living documents and trackers. The skill also covers Sheets automation and Gmail integration per its manifest. It fits teams whose operational truth lives in Google Workspace instead of a dedicated project database.8.4kinstalls 159Knowledge Opsknowledge-ops is an ECC agent skill for managing a multi-store knowledge base spanning local files, MCP memory, vector stores, Supabase, and Git repositories. The skill follows a live-workspace model: code stays in cloned repos, active execution context lives in GitHub, Linear, and working-context files, human notes can sit outside the repo, and cross-machine durable memory belongs in the knowledge base. Developers reach for knowledge-ops when they ask to save to a KB, sync knowledge, ingest documents, deduplicate entries, or query what the system already knows about a topic. It covers ingestion, structured storage, cross-system sync, deduplication, and retrieval tasks that exceed a single memory_store invocation.8.4kinstalls 160Workspace Surface Auditworkspace-surface-audit is an ECC read-only audit skill answering what a workspace and machine can currently do and what to add or enable next. It reviews active repositories, MCP servers, plugins, connectors, and environment surfaces including .env, .mcp.json, and plugin configuration without modifying files unless the user explicitly requests implementation. Developers reach for workspace-surface-audit when asking to set up Claude Code, recommend automation, choose plugins or MCP servers, or identify missing workflow layers. The skill compares official marketplace plugins against ECC-native coverage and determines whether a capability should be a skill, hook, agent, MCP server, or external connector. It is the ECC-native alternative to generic setup audit plugins.8.4kinstalls 161Llm Trading Agent Securityllm-trading-agent-security is an ECC skill for adding layered security controls when LLM agents receive authority to sign transactions or manage treasury operations. Developers reach for it when autonomous trading bots, DeFi agents, or treasury automation could move funds without human-in-the-loop safeguards. The skill targets key management boundaries, transaction approval flows, and agent permission minimization for high-risk financial automation. The published catalog entry notes an in-progress Japanese translation in some mirrors, but the English skill intent remains: prevent credential leakage and unauthorized signing before agents connect to wallets or exchange APIs. Use during security design reviews of LLM trading systems where a single prompt injection or tool misconfiguration could drain accounts. Pair with broader agent sandbox skills when shell, network, and secrets permissions overlap.8.4kinstalls 162Code Tourcode-tour is an Everything Claude Code skill that produces CodeTour-format walkthroughs instead of ad-hoc Markdown notes. The skill creates .tour JSON files under .tours/ with steps anchored to real files and line ranges, targeting specific reader personas for onboarding tours, architecture walkthroughs, PR reviews, and root-cause analysis paths. Developers reach for code-tour when a guided sequence serves better than a flat repository overview, such as ramping new engineers or explaining how a subsystem works with reusable artifacts. The skill only authors tour files and does not modify source code as part of the workflow.8.4kinstalls 163Gan Style Harnessgan-style-harness in affaan-m/ecc implements a GAN-inspired harness that separates generation from evaluation, creating an adversarial feedback loop for long-running application development. Inspired by Anthropic's harness design for long-running apps (March 24, 2026), the skill addresses a core failure mode: agents evaluating their own work tend toward pathological optimism, praising mediocre output and dismissing legitimate defects. The harness assigns distinct generator and evaluator roles so critique stays adversarial rather than self-congratulatory. Documented tools include Read, Write, Edit, Bash, Grep, Glob, and Task for filesystem and subprocess orchestration across agents. Developers reach for gan-style-harness when bootstrapping full applications from sparse prompts where iterative quality pressure beats one-shot codegen. The loop drives repeated generate-critique-revise cycles until evaluators accept production readiness. Triggers include multi-agent harness, generator-evaluator loop, adversarial code review, and long-running app scaffolding from minimal specs.8.4kinstalls 164Terminal OpsTerminal Ops is an ECC agent skill for evidence-priority repository execution when users ask to fix, debug, run commands, inspect the repo, or push narrow verified changes. It deliberately scopes narrower than generic coding guidance, focusing on command output, git state, test results, and documented verification rather than open-ended refactors. The skill stacks with related ECC skills: verification-loop for post-change checks, tdd-workflow when regression coverage is required, security-review when keys or external input are involved, github-ops for CI and PR status, and knowledge-ops to persist validated findings. Developers reach for Terminal Ops when tasks depend on real command execution and provable local verification before any push.8.4kinstalls 165Product Capabilityproduct-capability is an ECC agent skill that turns fuzzy product intent into explicit engineering constraints before coding starts. Instead of answering what to build in prose, the skill produces a capability plan listing invariants, interfaces, cross-service boundaries, and unresolved decisions developers must settle first. Teams use product-capability when a PRD, roadmap ticket, or stakeholder note exists but implementation assumptions remain implicit. The output is an ECC-native PRD-to-SRS lane that prevents agents from jumping straight into code without a shared contract.8.3kinstalls 166Agent Sortagent-sort is an ECC skill that builds an evidence-backed install plan for a specific repository by classifying ECC components with parallel repo-aware review passes. Instead of guessing what feels useful, agent-sort sorts skills, commands, rules, hooks, and extras into DAILY buckets for everyday use and LIBRARY buckets for occasional reference based on signals from the actual codebase. Developers use agent-sort when full ECC installs are too noisy, the repo stack is known, but nobody wants to manually audit dozens of skills and rules. The output is a trimmed ECC surface that loads only what the project needs, reducing agent context bloat and conflicting guidance. It pairs naturally with configure-ecc for installation after the sort plan is approved, and suits monorepos or polyglot projects where only a subset of ECC domains applies.8.3kinstalls 167Opensource Pipelineopensource-pipeline is a three-stage Claude Code skill from affaan-m/ecc that prepares private repositories for safe public release on GitHub. The pipeline forks a project and removes secrets, verifies the sanitized state, then packages required OSS files including CLAUDE.md, setup.sh, and README. Developers invoke it with `/opensource fork`, `/opensource verify`, or `/opensource package`, or by saying phrases like "open source this" or "make this public." The skill is built for teams that need a repeatable, agent-driven checklist instead of manual grep-and-hope secret audits before the first public push.8.3kinstalls 168Connections Optimizerconnections-optimizer is a growth networking skill from affaan-m/ecc (origin ECC) that reorganizes a developer's X and LinkedIn graph instead of treating outbound as a cold prospect list. The skill delivers review-first cleanup queues for X follows and LinkedIn connections, recommends new follows and connections matched to current priorities, identifies warm introduction paths, and drafts outreach in the user's voice across Apple Mail, X DM, and LinkedIn. Required inputs include current priorities, target roles or ecosystems, and platform choice (X, LinkedIn, or both). Developers reach for connections-optimizer when they want to prune low-signal follows, rebalance toward relevant peers and partners, or draft warm reconnection messages tied to network structure rather than bulk cold outreach.8.3kinstalls 169Csharp Testingcsharp-testing is an ECC agent skill for comprehensive .NET test development using a six-tool stack: xUnit as the test framework, FluentAssertions for readable assertions, NSubstitute or Moq for mocking, Testcontainers for real infrastructure in integration tests, WebApplicationFactory for ASP.NET Core integration tests, and Bogus for realistic test data. The skill covers Arrange-Act-Assert unit structure, test infrastructure setup, coverage review, and debugging flaky or slow tests. Developers reach for csharp-testing when writing new C# tests, auditing test quality, or standing up .NET test projects. Triggers include new test authoring, test infrastructure configuration, and flaky test investigation for .NET APIs and services.8.3kinstalls 170Project Flow Opsproject-flow-ops is an ECC skill that codifies a dual-track engineering workflow: GitHub remains the authoritative public record for issues, PRs, and roadmap visibility, while Linear holds only active, scheduled internal execution work. Teams adopt project-flow-ops to prevent duplicate ticket systems, reduce context switching, and keep external contributors aligned with GitHub while engineers run sprints and schedules in Linear. The published SKILL.md notes a Japanese translation in progress, but the described GitHub-plus-Linear operating model is the core deliverable. Engineering leads reach for project-flow-ops when scaling coordination across open-source visibility and private execution cadences.8.3kinstalls 171Hipaa Compliancehipaa-compliance is a thin, canonical entrypoint skill from affaan-m/ecc (version 1.0.0) for US healthcare compliance tasks. Instead of duplicating implementation detail, it delegates PHI/PII classification, audit logging, encryption, and leak prevention to healthcare-phi-compliance, healthcare-aware code review to healthcare-reviewer, and general auth, secrets, API, and deployment checks to security-review. Developers reach for hipaa-compliance when a task clearly involves HIPAA-regulated systems so the agent selects the right downstream compliance skills rather than applying generic security guidance. The skill stays intentionally minimal so healthcare-specific rules remain centralized in specialized companions.8.3kinstalls 172Automation Audit Opsautomation-audit-ops is an ECC operator skill that audits which automations are live, which jobs are broken, where overlap exists, and which connectors actually perform useful work—before rewriting anything. The first deliverable is an evidence-backed inventory plus keep, merge, cut, and fix recommendations covering jobs, hooks, MCP servers, wrappers, and connectors. It integrates related ECC skills such as workspace-surface-audit for connector and hook inventory, knowledge-ops for durable context alignment, github-ops for CI and scheduled workflows, ecc-tools-cost-audit for webhook fan-out and billing burn, and research-ops for platform truth checks. Developers reach for automation-audit-ops when automation sprawl, redundant MCP servers, or failing scheduled jobs obscure what is safe to change.8.3kinstalls 173Continuous Learning V2The continuous-learning-v2 skill implements an instinct-based learning system that observes Claude Code sessions through hooks and converts patterns into atomic instincts with confidence scores. Version 2.1 adds project-scoped instincts stored under ecc-homunculus project hashes to prevent cross-project contamination, detecting scope via git remote URL or repo path. Instincts can promote from project to global scope when seen in two or more projects. Commands expanded from four to six with promote and projects management alongside status, evolve, export, and import. The system evolves high-confidence instincts into full skills, commands, or agents over time. Activation covers configuring hooks, tuning confidence thresholds, reviewing instinct libraries, and managing project versus global scope. Universal patterns like input validation can remain global while framework-specific conventions stay project-local. Use when setting up automatic session learning, instinct extraction, or promoting learned behaviors into reusable agent assets.8.3kinstalls 174Research Opsresearch-ops is an ECC workflow skill that tells coding agents when and how to combine native research skills—exa-search, deep-research, market-research, lead-intelligence, and knowledge-ops—instead of replacing them. The skill fires on triggers like investigate, compare, who to contact, or latest information, and routes tasks to quick web discovery, multi-source cited synthesis, ranked recommendations, people or company enrichment, or persistent knowledge storage. Developers reach for research-ops when answers must reflect current public evidence plus local repository context, or when a one-off query should become a repeatable monitored workflow. It is a coordination layer over affaan-m/ecc’s research stack, not a standalone search API.8.3kinstalls 175Social Graph Rankersocial-graph-ranker is an ECC-origin Claude Code skill that applies a normalized weighted graph ranking layer across X and LinkedIn to score mutual followers and connections for warm-intro value. The skill ranks existing ties by intrinsic relationship strength, maps warm paths from your graph to a target ICP list, measures bridge value across first- and second-degree connections, and recommends warm introduction versus cold outreach. Developers reach for social-graph-ranker when they need the ranking engine itself—questions like who in my network can best introduce me—rather than full lead-generation pipelines handled by sibling ECC skills such as lead-intelligence or connections-optimizer. Use it standalone to inspect bridge math, compare introducer candidates, and prioritize network-aware distribution before sending messages.8.2kinstalls 176Email OpsEmail-ops is an operator workflow that turns an AI coding agent into a reliable mailbox coworker. It handles inbox triage, drafts replies, sends messages through the actual mail surface, and provides verifiable proof of what landed in Sent. The skill is deliberately not a generic writing tool; it enforces an evidence-first process that requires real client-side confirmation before claiming any message was sent. It automatically pulls in the right supporting ECC skills such as brand-voice for tone, investor-outreach for fundraising emails, or knowledge-ops to capture threads into durable context. Builders use it whenever the next concrete task is real email work rather than code or content generation.8.2kinstalls 177Defi Amm Securitydefi-amm-security is an ECC skill that codifies automated market maker smart-contract audit patterns for Uniswap-style, Curve-style, and Balancer-style protocols. It walks swap, mint, and burn logic through checks for flash-loan price manipulation, missing slippage guards, sandwich attacks, unsafe on-chain price oracles, reentrancy via checks-effects-interactions, and integer overflow handling. The embedded checklist verifies oracle-sourced pricing, flash-loan assumptions, slippage limits, reentrancy mutexes, and economic invariants across liquidity pools. Solidity developers reach for defi-amm-security when reviewing new AMM code, hardening swap functions, or testing protocol economics before mainnet deployment.8.1kinstalls 178Ecc Tools Cost Auditecc-tools-cost-audit is a Claude Code skill from affaan-m/ecc focused on the ECC Tools GitHub App sibling repository. The workflow investigates runaway costs, quota bypasses, duplicate PRs, and premium model routing for free users by analyzing prompt input tokens and quantifying compute efficiency. It composes with ECC-native skills such as autonomous-loops for cross-webhook billing audits and agentic-engineering for implementation review, but stays a targeted operator pass—not generic billing or full repo review. Developers reach for ecc-tools-cost-audit when ECC Tools webhooks, queues, or retries appear to burn tokens or spawn excess pull requests.8.1kinstalls 179Messages OpsMessages-ops is an ECC agent skill that implements an evidence-first workflow for live messaging surfaces including iMessage, social DMs, and browser-gated inboxes. The SKILL.md defines a four-step process to resolve the exact thread, read inbound messages before drafting replies, recover recent one-time codes with focused search, and report verifiable evidence naming the source, sender, time window, and status. Developers reach for messages-ops when an AI coding agent must check local texts, inspect a DM thread, or find an MFA code delivered to a messaging app instead of guessing or conflating mailbox work with DM retrieval. Guardrails block unverified retrieval claims, document four adjacent ECC skills for handoff, and require structured OUTPUT with SOURCE, RESULT, and STATUS blocks. Install with npx skills add affaan-m/everything-claude-code --skill messages-ops.8.1kinstalls 180Unified Notifications Opsunified-notifications-ops is an ECC-native workflow skill for developers drowning in fragmented alerts rather than missing notifications entirely. It merges GitHub Issues, PRs, reviews, comments, and CI events with Linear issue status changes, local hooks, desktop alerts, chat, and email into a single operator interface with explicit severity levels, owners, routing rules, and next actions. Developers reach for unified-notifications-ops when CI failures, review requests, and operator events scatter across channels and generate noise instead of actionable signal. The skill assumes hooks, MCP servers, or connected tools already exist but lack a consistent notification strategy. It prioritizes deduplication, escalation paths, and inbox collapse over adding yet another alert source.8.1kinstalls 181Hookify Ruleshookify-rules is an ECC skill for writing Hookify rules—markdown files with YAML frontmatter that define regex patterns to watch and messages to show when patterns match during Claude sessions. Rules are stored as .claude/hookify.{rule-name}.local.md files. Frontmatter fields include name, enabled, event (bash|file|stop|prompt|all), and pattern (regex), followed by the message body shown on trigger with markdown formatting, warnings, and suggestions. Developers reach for hookify-rules when they want automatic guardrails against dangerous bash commands, risky file edits, or stop/prompt patterns without building a full pre-commit hook stack. The skill covers rule file format, event types, and rule-driven workflows for event-driven execution inside Claude. Use when creating, editing, or managing custom Hookify rules that surface guidance the moment matching code or commands appear.8.1kinstalls 182Evm Token Decimalsevm-token-decimals is an ECC skill at version 1.0.0 that stops silent decimal-mismatch bugs where ERC-20 balances or USD values appear correct but are wrong by orders of magnitude. The skill queries decimals() at runtime, caches results by chain_id and token_address, and applies decimal-safe math in Python, TypeScript, or Solidity. Developers reach for evm-token-decimals when reading on-chain balances, computing fiat values, comparing token amounts across EVM chains, or handling bridged assets in portfolio trackers, bots, and DeFi aggregators. The workflow never assumes stablecoins share the same decimals and covers bridged token precision drift. Runtime inspection plus cached chain-aware normalization is the core pattern.8.1kinstalls 183Docker Patternsname docker-patterns description Docker and Docker Compose patterns for local development container security networking volume strategies and multi-service orchestration Use when setting up containerized development environments or reviewing Docker configurations metadata origin ECC Docker Patterns Docker and Docker Compose best practices for containerized development When to Activate Setting up Docker Compose for local development Designing multi-container architectures Troubleshooting container networking or volume issues Reviewing Dockerfiles for security and size Migrating from local dev to containerized workflow Docker Compose for Local Development Standard Web App Stack yaml docker-compose yml services app build context target dev Use dev stage of multi-stage Dockerfile ports 3000 3000 volumes app Bind mount for hot reload app node_modules Anonymous volume preserves container deps environment DATABASE_URL postgres postgres postgres db 5432 app_dev REDIS_URL redis redis 6379 0 NODE_ENV development depends_on db condition service_healthy redis condition service_started command npm run dev db image postgres 16-alpine ports 5432 5432 environment POSTGRES_USER postgres POSTGRES_P.7.9kinstalls 184Django PatternsThis skill provides production-grade Django architecture patterns for developers building scalable web applications and REST APIs. It covers Django project structure (split settings for dev/prod/test), ORM best practices (custom QuerySets, managers, select_related/prefetch_related for N+1 prevention), Django REST Framework patterns (serializers with validation, ViewSets with custom actions), service layer design for business logic separation, caching strategies (view-level, template fragment, low-level QuerySet caching), signals for event-driven workflows, custom middleware, and performance optimization through database indexing and bulk operations. Developers invoke this when structuring new Django projects, designing DRF endpoints, optimizing database queries, or implementing caching and middleware.7.9kinstalls 185Postgres PatternsThe postgres-patterns skill is a PostgreSQL quick reference for query optimization, schema design, indexing, and security based on Supabase best practices. Agents match index types to query patterns including B-tree equality and range filters, composite indexes, GIN for JSONB and full text, and BRIN for time series. Data type guidance favors bigint IDs, text strings, timestamptz, numeric money, and boolean flags over weaker alternatives. Patterns cover composite index column order, covering indexes with INCLUDE, partial indexes, optimized RLS policies wrapping auth.uid in SELECT, UPSERT, cursor pagination, and queue processing with FOR UPDATE SKIP LOCKED. Anti-pattern SQL finds unindexed foreign keys, slow pg_stat_statements queries, and table bloat. A configuration template sets connection limits, work_mem, timeouts, pg_stat_statements, and public schema revokes. Activation triggers include writing SQL, designing schemas, troubleshooting slow queries, implementing RLS, and connection pooling. Related tools include database-reviewer, backend-patterns, and database-migrations skills.7.9kinstalls 186Python PatternsThis skill provides Python-specific design patterns and best practices for developers writing Pythonic code. It covers protocols for duck typing with type hints, dataclasses for DTOs, context managers for resource handling, generators for memory efficiency, decorators for cross-cutting concerns, async/await patterns for concurrent I/O, advanced type hints including generics and ParamSpec, dependency injection, package organization, custom exceptions, property decorators, and functional programming utilities. Developers invoke this when structuring Python projects, designing APIs, implementing async systems, or refactoring codebases to use idiomatic Python patterns. The skill extends common design principles with Python-specific idioms like frozen dataclasses, contextlib decorators, and asyncio.gather for concurrent execution.7.8kinstalls 187Gateguardgateguard is a Claude Code skill from affaan-m/ecc implementing GateGuard, a PreToolUse hook that forces investigation before any file edit. Instead of LLM self-evaluation prompts like are you sure, gateguard demands concrete facts—actual paths, schema fields, date formats, and import dependencies—so the act of reading creates awareness self-checks never provide. Developers activate it on multi-module codebases, data files with strict schemas, and teams requiring AI edits to match existing patterns. The skill targets workflows where agents tend to hallucinate file layout or API shapes, turning pre-edit investigation into a hard gate rather than a soft suggestion.7.8kinstalls 188Frontend SlidesFrontend Slides generates single-file HTML presentations with inline CSS and JavaScript, designed to run entirely in the browser without dependencies. Developers use it when building talk decks, pitch presentations, or converting PowerPoint files to web format. The skill enforces viewport-fit constraints so every slide displays without scrolling across devices from mobile to 1920x1080. It guides users through visual style exploration with preview files rather than abstract questionnaires, supports PPT/PPTX extraction via python-pptx, and includes keyboard/touch/wheel navigation, reveal animations, and reduced-motion support. The workflow covers content gathering, style discovery through mood-based presets, implementation with semantic HTML, accessibility validation, and cross-platform file delivery with platform-specific openers.7.7kinstalls 189Video EditingAI-assisted video editing workflows for cutting structuring and augmenting real footage Covers the full pipeline from raw capture through FFmpeg Remotion ElevenLabs fal ai and final polish in Descript or CapCut Use when the user wants to edit video cut footage create vlogs or build video content name video-editing description AI-assisted video editing workflows for cutting structuring and augmenting real footage Covers the full pipeline from raw capture through FFmpeg Remotion ElevenLabs fal ai and final polish in Descript or CapCut Use when the user wants to edit video cut footage create vlogs or build video content Video Editing AI-assisted editing for real footage When to Activate User wants to edit cut or structure video footage Turning long recordings into short-form content Building vlogs tutorials or demo videos from raw capture Adding overlays subtitles music or voiceover to existing video Reframing video for different platforms YouTube TikTok Instagram User says edit video cut this footage make a vlog or video workflow Core Thesis AI video editing is useful when you7.7kinstalls 190Api DesignThis skill provides comprehensive REST API design patterns for building production-grade HTTP APIs. Developers invoke it when designing new endpoints, implementing pagination and filtering, standardizing error responses, or planning API versioning strategies. It covers resource URL structuring with plural nouns and kebab-case, HTTP method semantics with idempotency rules, status code selection (200/201/204/400/401/403/404/422/429/500/503), pagination strategies (offset-based for small datasets vs cursor-based for scale), filtering with bracket notation for operators, sorting with prefix syntax, authentication patterns, and rate limiting headers. The skill includes TypeScript/Next.js, Python/Django REST Framework, and Go net/http implementations, plus a pre-launch checklist for validation, authorization, error formatting, and documentation.7.6kinstalls 191Tdd WorkflowThis skill enforces test-driven development principles where developers write tests before implementation code. It provides structured workflows for creating user journeys, generating test cases, implementing minimal code to pass tests, and refactoring while maintaining 80%+ coverage. The skill covers unit tests for functions and components, integration tests for APIs and database operations, and E2E tests using Playwright for critical user flows. Developers follow a seven-step cycle: write user journeys, generate test cases, run failing tests, implement code, verify passing tests, refactor, and check coverage. The skill includes patterns for Jest/Vitest unit tests, Next.js API integration tests, and Playwright E2E tests, with mocking strategies for Supabase, Redis, and OpenAI.7.6kinstalls 192Java Coding StandardsJava coding standards for Spring Boot and Quarkus services naming immutability Optional usage streams exceptions generics CDI reactive patterns and project layout Automatically applies framework-specific conventions name java-coding-standards description Java coding standards for Spring Boot and Quarkus services naming immutability Optional usage streams exceptions generics CDI reactive patterns and project layout Automatically applies framework-specific conventions origin ECC Java Coding Standards Standards for readable maintainable Java 17 code in Spring Boot and Quarkus services When to Use Writing or reviewing Java code in Spring Boot or Quarkus projects Enforcing naming immutability or exception handling conventions Working with records sealed classes or pattern matching Java 17 Reviewing use of Optional streams or generics Structuring packages and project layout QUARKUS Working with CDI scopes Panache entities or reactive pipelines How It Works Framework Detection Before applying standards determine the framework from the build file Build file contains quarkus apply QUARKUS conventions Build file contains spring-boot apply SPRING conventions Neither detected apply shared con.7.4kinstalls 193Django SecurityDjango security best practices authentication authorization CSRF protection SQL injection prevention XSS prevention and secure deployment configurations name django-security description Django security best practices authentication authorization CSRF protection SQL injection prevention XSS prevention and secure deployment configurations origin ECC Django Security Best Practices Comprehensive security guidelines for Django applications to protect against common vulnerabilities When to Activate Setting up Django authentication and authorization Implementing user permissions and roles Configuring production security settings Reviewing Django application for security issues Deploying Django applications to production Core Security Settings Production Settings Configuration python settings production py import os DEBUG False CRITICAL Never use True in production ALLOWED_HOSTS os environ get ALLOWED_HOSTS split Security headers SECURE_SSL_REDIRECT True SESSION_COOKIE_SECURE True CSRF_COOKIE_SECURE True SECURE_HSTS_SECONDS 31536000 1 year SECURE_HSTS_INCLUDE_SUBDOMAINS True SECURE_HSTS_PRELOAD True SECURE_CONTENT_TYPE_NOSNIFF True SECURE_BROWSER_XSS_FILTER True X_FRAME_OPTIONS DENY HTTPS.7.3kinstalls 194Python Testingname python-testing description Python testing best practices using pytest including fixtures parametrization mocking coverage analysis async testing and test organization Use when writing or improving Python tests metadata origin ECC globs py pyi Python Testing This skill provides comprehensive Python testing patterns using pytest as the primary testing framework Testing Framework Use pytest as the testing framework for its powerful features and clean syntax Basic Test Structure python def test_user_creation Test that a user can be created with valid data user User name Alice email alice example com assert user name Alice assert user email alice example com assert user is_active is True Test Discovery pytest automatically discovers tests following these conventions Files test_ py or _test py Functions test_ Classes Test without __init__ Methods test_ Fixtures Fixtures provide reusable test setup and teardown python import pytest from sqlalchemy import create_engine from sqlalchemy orm import sessionmaker pytest fixture def db_session Provide a database session for tests engine create_engine sqlite memory Session sessionmaker bind engine session Session Setup Base metadata create_.7.3kinstalls 195Frontend Design Directionfrontend-design-direction is an ECC community skill salvaged from stale PR #1659 that guides developers to define aesthetic principles and a coherent design language before implementing web pages, apps, dashboards, artifacts, or components. The skill targets work where UI must feel purposeful and appropriate to the product domain, not merely functional. ECC intentionally does not rebundle Anthropic's canonical frontend-design skill; developers install that from anthropics/skills for the official upstream version. Reach for frontend-design-direction when starting a new frontend surface and you need domain-appropriate visual direction, typography, spacing, and polish decisions documented before coding begins.7.3kinstalls 196Prompt OptimizerThe prompt-optimizer skill listing in this catalog cache is a Japanese translation stub for the everything-claude-code prompt-optimizer upstream skill rather than a complete English SKILL.md. Front matter states translation is in progress and points agents to the original English file at the upstream everything-claude-code prompt-optimizer path for full optimization workflows. Agents should acknowledge the documentation gap, avoid inventing prompt rewrite pipelines not present locally, and fetch or request the upstream English skill before promising production prompt optimization. The stub still signals intent around tuning agent prompts for clarity, token efficiency, and task success, but this cached copy alone does not document commands, evaluation loops, or before-and-after examples. Route detailed optimization work to verified upstream English documentation before executing automation or claiming measured improvements. Treat measured prompt changes as unverified until upstream English examples are loaded.7.2kinstalls 197Springboot SecuritySpring Security best practices for authn authz validation CSRF secrets headers rate limiting and dependency security in Java Spring Boot services name springboot-security description Spring Security best practices for authn authz validation CSRF secrets headers rate limiting and dependency security in Java Spring Boot services origin ECC Spring Boot Security Review Use when adding auth handling input creating endpoints or dealing with secrets When to Activate Adding authentication JWT OAuth2 session-based Implementing authorization PreAuthorize role-based access Validating user input Bean Validation custom validators Configuring CORS CSRF or security headers Managing secrets Vault environment variables Adding rate limiting or brute-force protection Scanning dependencies for CVEs Authentication Prefer stateless JWT or opaque tokens with revocation list Use httpOnly Secure SameSite Strict cookies for sessions Validate tokens with OncePerRequestFilter or resource server java Component public class JwtAuthFilter extends OncePerRequestFilter private final JwtService jwtService public JwtAuthFilter JwtService jwtService this jwtService jwtService Override protected void doFilterInternal.7.2kinstalls 198Article WritingTurn raw notes, transcripts, or research into polished long-form articles, blog posts, guides, and newsletters that match a distinctive brand or founder voice.7kinstalls 199Everything Claude Code ConventionsEverything Claude Code Conventions teaches development patterns for the affaan-m/everything-claude-code JavaScript repository. Use when contributing to ECC projects or learning hybrid module layout, conventional commit formatting, and test organization patterns.7kinstalls 200Cpp Coding StandardsC coding standards based on the C Core Guidelines isocpp github io Use when writing reviewing or refactoring C code to enforce modern safe and idiomatic practices name cpp-coding-standards description C coding standards based on the C Core Guidelines isocpp github io Use when writing reviewing or refactoring C code to enforce modern safe and idiomatic practices origin ECC C Coding Standards C Core Guidelines Comprehensive coding standards for modern C C 17 20 23 derived from the C Core Guidelines https isocpp github io CppCoreGuidelines CppCoreGuidelines Enforces type safety resource safety immutability and clarity When to Use Writing new C code classes functions templates Reviewing or refactoring existing C code Making architectural decisions in C projects Enforcing consistent style across a C codebase Choosing between language features e g enum vs enum class raw pointer vs smart pointer When NOT to Use Non-C projects Legacy C codebases that cannot adopt modern C features Embedded bare-metal contexts where specific guidelines conflict with hardware constraints adapt selectively Cross-Cutting Principles These themes7kinstalls 201Everything Claude Codeeverything-claude-code is a repo-specific agent skill generated from affaan-m/everything-claude-code on 2026-03-20 that encodes development patterns and workflows for a JavaScript project using conventional commits. Developers invoke it with the default prompt to follow repository conventions and workflows instead of generic coding advice. The skill supports implicit invocation so Claude applies ECC patterns automatically during edits, reviews, and commits. Reach for everything-claude-code when contributing to or extending the everything-claude-code project and need agents to mirror its established architecture and commit message rules.7kinstalls 202Hermes Importshermes-imports is an ECC skill for turning stable Hermes operator loops into public-ready Claude Code skills. Hermes acts as the local operator shell while ECC holds the reusable workflow layer; this skill moves repeatable patterns across that boundary without copying private state. The workflow follows a six-step conversion pattern: identify the loop, strip private inputs, rewrite paths as repo-relative examples, add When To Use guidance, define output requirements, and run a secret scan before PR. A sanitization checklist targets absolute paths, tokens, OAuth files, CRM or health data, and raw private logs. The output contract returns a candidate skill name, sanitized summary, required public inputs, removed private inputs, remaining risks, and files to create. Developers reach for hermes-imports when a Hermes job has repeated enough to merit ECC publication.7kinstalls 203Fastapi PatternsFastAPI Patterns is an ECC community skill that treats a FastAPI app as a thin HTTP layer over explicit routers, schemas, dependencies, and database access modules. It activates when building or reviewing FastAPI services—splitting routers from Pydantic request/response models, wiring async endpoints that call databases or external services, and adding authentication, authorization, OpenAPI documentation, tests, and deployment settings. Developers reach for it instead of assembling patterns from scattered docs when scaffolding a new API or reviewing a FastAPI PR for production risks. The skill emphasizes copy-pasteable, production-oriented examples covering dependency injection, async I/O, and security rather than toy CRUD snippets.6.9kinstalls 204Market ResearchGenerate sourced, decision-oriented market research, competitor analysis, and investor due diligence that directly informs product and business choices.6.9kinstalls 205Android Clean ArchitectureThis skill defines clean architecture patterns for Android and Kotlin Multiplatform (KMP) projects, covering module boundaries, dependency inversion, and data flow across domain, data, and presentation layers. Developers use it when scaffolding new Android or KMP projects, implementing UseCase and Repository abstractions, or wiring up Room/SQLDelight databases with Ktor network clients. Key workflows include defining pure-Kotlin domain models with no framework annotations, implementing repository classes that coordinate local and remote data sources via mapper extension functions, and configuring dependency injection with either Koin (KMP-compatible) or Hilt (Android-only). The skill also covers Gradle convention plugins to reduce build file duplication across modules, sealed error types for propagating failures through Result or custom Try types, and explicit anti-patterns such as placing business logic in ViewModels or exposing database entities to the UI layer.6.9kinstalls 206Strategic CompactSuggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction name strategic-compact description Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction Strategic Compact Skill Suggests manual compact at strategic points in your workflow rather than relying on arbitrary auto-compaction When to Activate Running long sessions that approach context limits 200K tokens Working on multi-phase tasks research plan implement test Switching between unrelated tasks within the same session After completing a major milestone and starting new work When responses slow down or become less coherent context pressure Why Strategic Compaction Auto-compaction triggers at arbitrary points Often mid-task losing important context No awareness of logical task boundaries Can interrupt complex multi-step operations Strategic compaction at logical boundaries After exploration before execution Compact research context keep implementation plan After completing a milestone Fresh start for next phase Before major context shifts Clear exploration context before diff.6.9kinstalls 207Vite Patternsvite-patterns is an ECC-origin skill covering Vite 8+ development server and build toolchain patterns. It activates for vite.config.ts or vite.config.js setup, .env and environment variables, dev-server proxy to API backends, build output optimization for chunks and minification, build.lib library publishing, dependency pre-bundling, and CJS/ESM interop troubleshooting. Developers also use it when selecting plugin order or debugging HMR and build errors. The skill replaces trial-and-error Vite configuration with repeatable patterns for modern frontend and library packages.6.8kinstalls 208Agent Architecture Auditagent-architecture-audit is an ECC diagnostic workflow (origin: oh-my-agent-check) for agent and LLM applications. It audits a 12-layer agent stack for wrapper-layer regressions, stale memory pollution, tool discipline failures, hidden retry loops, and transport or rendering mutations that mask root causes. Required before production when tool calls, memory, or multi-step workflows ship—or when the same model works in a playground but breaks inside wrappers. Tools used include Read, Write, Edit, Bash, Grep, and Glob. Developers reach for agent-architecture-audit after 15+ minutes of blind agent debugging or when users report unstable tools and degraded agent behavior.6.7kinstalls 209Liquid Glass DesignThe liquid-glass-design skill documents Apple's iOS 26 Liquid Glass design system for SwiftUI, UIKit, and WidgetKit. Liquid Glass is a dynamic material that blurs content behind it, reflects surrounding color and light, and reacts to touch or pointer input. Core SwiftUI patterns start with glassEffect on padded views, then customize regular variants with tint colors, interactive modifiers, and rect or capsule shapes. Button styles include glass and glassProminent for hierarchy. Performance guidance requires wrapping multiple glass views in GlassEffectContainer with spacing that controls merge distance and morphing behavior. Advanced patterns use glassEffectUnion with namespaces to blend sibling elements and glassEffectID for smooth appear and disappear transitions. Activate when updating apps to iOS 26 design language, replacing legacy blur materials, building glass toolbars or cards, or applying Liquid Glass effects to widgets. The skill covers migration from older material APIs and widget integration constraints for consistent glass presentation across Apple platforms.6.7kinstalls 210Springboot TddDesarrollo guiado por pruebas para Spring Boot usando JUnit 5 Mockito MockMvc Testcontainers y JaCoCo Usar al agregar funcionalidades corregir bugs o refactorizar name springboot-tdd description Desarrollo guiado por pruebas para Spring Boot usando JUnit 5 Mockito MockMvc Testcontainers y JaCoCo Usar al agregar funcionalidades corregir bugs o refactorizar origin ECC Flujo de Trabajo TDD en Spring Boot Orientación TDD para servicios Spring Boot con 80 de cobertura unit integración Cuándo Usar Nuevas funcionalidades o endpoints Correcciones de bugs o refactorizaciones Agregar lógica de acceso a datos o reglas de seguridad Flujo de Trabajo 1 Escribir pruebas primero deben fallar 2 Implementar el código mínimo para que pasen 3 Refactorizar con pruebas en verde 4 Exigir cobertura con JaCoCo Pruebas Unitarias JUnit 5 Mockito java ExtendWith MockitoExtension class class MarketServiceTest Mock MarketRepository repo InjectMocks MarketService service Test void createsMarket CreateMarketRequest req new CreateMarketRequest name desc Instant now List of cat when repo save any thenAnswer inv inv getArgument 0 Market result service create req assertThat result name isEqualTo name verify6.7kinstalls 211Redis Patternsredis-patterns is an ECC backend reference skill covering Redis data structures, atomic workflows, and production configuration for common application workloads. The skill explains when to use strings, hashes, lists, sets, sorted sets, and streams, and how multi-step operations require Lua scripts, MULTI/EXEC transactions, or explicit synchronization beyond single-command atomicity. Developers reach for redis-patterns when adding caching layers, rate limiters, distributed locks, session stores, or pub/sub and Redis Streams messaging, including pooling, eviction, and clustering setup. The included data-structure cheat sheet maps use cases to Redis types with concrete command examples for backend services.6.7kinstalls 212Error Handlingerror-handling is a skill from affaan-m/ecc that standardizes robust error patterns for production applications in TypeScript, Python, and Go. The skill promotes failing fast at boundaries, using typed errors instead of string messages, separating user-facing text from developer logs, and documenting every client-visible error code as part of the API contract. Developers activate error-handling when designing exception hierarchies, adding retry or circuit-breaker logic for unreliable dependencies, reviewing missing endpoint error handling, or debugging cascade failures and silent catch blocks. Five core principles anchor every recommendation so errors surface early and never disappear inside empty catch blocks.6.7kinstalls 213Database MigrationsThis skill provides structured patterns and checklists for executing safe, reversible database schema changes in production systems. Developers use it when adding or removing columns, creating indexes, renaming fields, or running data backfills without causing downtime or table locks. The skill covers PostgreSQL-specific techniques such as CREATE INDEX CONCURRENTLY and batch UPDATE loops, plus ORM-specific workflows for Prisma, Drizzle, Django, and golang-migrate. A core workflow is the expand-contract pattern, which separates schema changes, data backfills, and column removal across multiple deployments. Anti-pattern tables document common mistakes - such as adding NOT NULL columns without defaults or mixing DDL and DML - alongside safer alternatives grounded in production-scale concerns like 10M-row tables and lock contention.6.7kinstalls 214Laravel SecurityApply Laravel security best practices for authentication, input validation, CSRF protection, rate limiting, and secure configuration.6.6kinstalls 215Mysql Patternsmysql-patterns is an ECC-origin agent skill that encodes production MySQL and MariaDB conventions for schema design, query tuning, indexing strategy, transactional safety, replication setup, and connection pool configuration. It activates when developers design tables, write slow queries, plan indexes, or configure pools in MySQL-compatible backends. The skill distills repeatable patterns so agents propose normalized schemas, efficient indexes, and pool settings without rediscovering common pitfalls. Developers reach for mysql-patterns when building or refactoring SaaS APIs, admin dashboards, or services backed by MySQL or MariaDB and need opinionated guidance instead of generic SQL snippets.6.6kinstalls 216Windows Desktop E2ewindows-desktop-e2e is an ECC skill for end-to-end testing Windows native desktop applications with pywinauto backed by Windows UI Automation. It covers WPF, WinForms, Win32 and MFC, and Qt 5.x through 6.x, rating AutomationId reliability per framework in a comparison table. Setup installs pywinauto, pytest, pytest-html, Pillow, and pytest-timeout on Python 3.8+ Windows hosts and recommends Accessibility Insights for inspecting the UIA element tree. The skill teaches page-object models, AutomationId-first locator strategy, wait patterns, per-step trace artifacts, flaky-test skips, and sandboxed APPDATA isolation via subprocess fixtures. Qt guidance documents QT_ACCESSIBILITY=1 for older releases and keyboard fallbacks when UIA ValuePattern is missing. CI integration targets GitHub Actions windows-latest runners. Developers reach for windows-desktop-e2e when Playwright cannot drive a native .exe GUI and desktop automation tests need stable UIA selectors or CI wiring.6.6kinstalls 217Django TddDjango TDD skill documents test-driven development for Django applications using pytest, factory_boy, and Django REST Framework. It walks the Red-Green-Refactor cycle from failing tests through green implementations to refactors that stay green. Setup covers pytest.ini with reuse-db and nomigrations, a fast in-memory SQLite test settings module, and conftest fixtures for users, authenticated clients, and DRF API clients. Factory Boy factories generate Users, Categories, and Products with sequences, fuzzy pricing, and post-generation tags. Model tests cover validation, managers, and stock rules; view tests assert auth redirects and POST flows; serializer tests verify price and stock constraints; DRF API tests exercise list, create, update, delete, filter, and search endpoints. Mocking patterns patch external payment gateways and assert email outbox delivery. Integration tests model full checkout flows. Best practices favor factories over manual setup, descriptive names, edge cases, fast settings, and coverage goals from 80 percent overall to 90 percent on models and services.6.6kinstalls 218Security ScanThe security-scan skill audits Claude Code project configuration using AgentShield (ecc-agentshield) to detect security vulnerabilities, misconfigurations, and injection risks across the .claude/ directory. It inspects CLAUDE.md for hardcoded secrets and prompt-injection patterns, settings.json for overly permissive allowlists and missing deny lists, mcp.json for risky MCP servers and hardcoded environment secrets, hooks for command injection via interpolation and data exfiltration, and agents/*.md for unrestricted tool access and missing model specifications. Run npx ecc-agentshield scan with optional --path, --min-severity, --format json, markdown, or html, and --fix for safe auto-remediation of secrets and wildcard permissions. The --opus pipeline runs attacker, defender, and auditor agents for deeper analysis when ANTHROPIC_API_KEY is set. Use agentshield init to bootstrap scoped settings.json, CLAUDE.md, and mcp.json placeholders, and add affaan-m/agentshield@v1 to CI with fail-on-findings. Severity grades span A (90-100 safe) through F (0-39 critical). Invoke on new Claude Code setup, settings changes, pre-commit reviews, repository onboarding, or periodic hygiene checks.6.6kinstalls 219Laravel PatternsGet production-grade Laravel architecture patterns that enforce clean boundaries between controllers, services, models, and jobs.6.6kinstalls 220Continuous LearningContinuous Learning v1 is a deprecated Stop-hook skill for Claude Code that evaluates sessions at end-of-session to extract reusable patterns and save them to ~/.claude/skills/learned/. It triggers only when a session reaches a configurable minimum message count (default 10+), then detects patterns across five categories: error resolution, user corrections, workarounds, debugging techniques, and project-specific conventions. Configuration is managed via config.json, and the hook is registered in ~/.claude/settings.json under the Stop event. Compared to v2, this version observes only at session end (not per-tool-use), stores skills globally without project scoping, and lacks confidence scoring or instinct-based atomic behavior tracking. Developers maintaining older installs or needing the simpler extraction flow can keep v1, but new projects should use continuous-learning-v2.6.6kinstalls 221Make Interfaces Feel Bettermake-interfaces-feel-better in affaan-m/ecc (origin ECC) guides agents to elevate generic UI components into polished, tactile interfaces that feel finished and delightful rather than template-default. The skill targets micro-interactions, visual hierarchy, spacing, motion, and affordances that separate prototype UIs from production-quality surfaces. A Japanese translation of the skill file is in progress; the English source remains the authoritative workflow reference. Developers reach for make-interfaces-feel-better when a React or web UI technically works but feels flat, when design review asks for more polish, or when shipping a user-facing screen that needs perceived quality improvements without a full redesign. Triggers include make UI feel better, polish components, improve tactile feedback, and refine generic shadcn or Tailwind layouts.6.6kinstalls 222Production Auditproduction-audit is an ECC agent skill that performs a local production-readiness audit to surface issues that could break in production while keeping all repository data on the developer machine. The audit is designed for teams that want pre-deploy confidence without uploading code or configuration to third-party scanning services. Developers reach for production-audit when preparing a release, hardening an AI-assisted codebase, or validating that environment, dependency, and operational gaps are caught before traffic hits production. The skill originates from the Everything Claude Code collection and complements external CI checks with an offline-first review pass.6.6kinstalls 223E2e TestingThe e2e-testing skill documents comprehensive Playwright patterns for stable, fast, maintainable end-to-end suites. It prescribes tests/e2e folder organization by auth, features, and API areas plus fixtures and playwright.config.ts at the repo root. Page Object Model examples encapsulate locators such as data-testid search inputs and item cards with goto, search, and count helpers that wait for network responses instead of arbitrary sleeps. Configuration covers fullyParallel runs, CI retries at two, forbidOnly in CI, HTML and JUnit reporters, trace on-first-retry, screenshot and video on failure, multi-browser projects including mobile Chrome, and webServer dev startup with reuseExistingServer locally. Flaky test guidance covers quarantine via test.fixme or conditional test.skip, repeat-each debugging, and fixes for race conditions, network timing, and animation stability using auto-wait locators and waitForResponse. Artifact sections cover screenshots, traces, and retained failure videos. CI/CD includes a GitHub Actions workflow with playwright install --with-deps and artifact upload. Additional recipes cover wallet provider mocking and guarded financial flow tests that skip on p.6.6kinstalls 224Motion Patternsmotion-patterns is a Claude Code skill from affaan-m/ecc (Everything Claude Code) that delivers reusable animation patterns for React and Next.js interfaces. The skill covers accessible motion defaults—durations, spring configurations, and easing functions—so developers avoid inconsistent or performance-heavy custom animations. Developers reach for motion-patterns when building dashboards, marketing pages, or interactive components that need polished enter, exit, and layout transitions aligned with modern motion libraries. The published readme notes Japanese translation is in progress, but the skill origin and description target English-language React and Next.js animation workflows with accessibility considerations baked in.6.6kinstalls 225Motion Foundationsmotion-foundations is an Everything Claude Code (ECC) skill for setting animation baselines in React and Next.js apps. It guides developers to define performant, accessible motion patterns before importing animation libraries, reducing jank and a11y regressions. Teams reach for motion-foundations when starting UI motion work or standardizing transitions across components. The published readme notes Japanese translation is in progress; the English skill file in the ECC repository holds the full motion guidance referenced by the stub.6.5kinstalls 226Motion Advancedmotion-advanced is a skill from affaan-m/ecc in the Everything Claude Code collection, focused on advanced Motion-powered interactions for React and Next.js UIs. The published readme notes the public file is a Japanese translation placeholder and points to the original English SKILL.md for full guidance on drag-and-drop, gesture handling, text reveals, SVG path animations, and imperative animation control. Developers reach for motion-advanced when standard fade or slide transitions are insufficient and interfaces need physics-based dragging, gesture-driven states, or path-based SVG motion. The skill complements baseline motion-ui guidance by covering higher-complexity interaction patterns during active frontend implementation.6.5kinstalls 227Deployment Patternsdeployment-patterns teaches production deployment workflows and CI/CD best practices for web applications. Use it when setting up CI/CD pipelines, Dockerizing an application, planning rolling, blue-green, or canary strategies, implementing health checks and readiness probes, or preparing a production release. The skill compares rolling, blue-green, and canary rollouts with pros, cons, and when each fits. It provides multi-stage Dockerfiles for Node.js, Go, and Python with non-root users, pinned tags, HEALTHCHECK instructions, and layer caching. GitHub Actions pipelines cover test, build, and deploy stages with lint, typecheck, coverage, image push to ghcr.io, and environment-gated production deploys. Health check endpoints include simple and detailed database, Redis, and external API probes plus Kubernetes liveness, readiness, and startup probes. Twelve-factor environment configuration uses zod validation at startup. Rollback strategies cover kubectl rollout undo, Vercel rollback, Railway redeploy, and migration compatibility. A production readiness checklist spans application tests, infrastructure limits, monitoring alerts, security headers, and operations runbooks.6.5kinstalls 228Blender Motion State InspectionBlender Motion State Inspection is an ECC agent skill for diagnosing Blender character and animation problems when visual screenshots are insufficient. It compares rendered evidence against structured facts such as bone hierarchies, bounding boxes, foot-ground contacts, and facing direction vectors to detect twisted, mirrored, flattened, or offset rigs and foot-sliding in retargeted motion. Developers reach for Blender Motion State Inspection when imported avatars, armatures, or retargeted animations may not match expected poses or when model-vs-motion alignment must be verified programmatically. The skill uses Read, Write, Edit, Bash, Grep, and Glob tools to inspect .blend assets and produce factual diagnostics agents can act on.6.5kinstalls 229Django CeleryDjango Celery is an ECC agent skill for adding asynchronous task processing to Django projects using Celery with Redis or RabbitMQ brokers. It walks through installing Celery and Redis via pip, defining `@shared_task` functions, starting workers with `celery -A myapp worker -l info`, and enqueueing work with `.delay()` for immediate async execution or crontab-based schedules for periodic jobs like daily reports and cleanup tasks. Developers reach for Django Celery when email sending, heavy processing, recurring maintenance, external API calls, or multi-step workflows should not block HTTP request threads. The skill covers task queuing, worker management, error handling, and broker integration patterns standard in production Django deployments.6.5kinstalls 230Homelab Network Setuphomelab-network-setup is a community ECC skill for developers planning or redesigning home and small-lab networks that must scale without tearing everything down later. It separates device roles—from modem through gateway, switch, and access points—and designs IP ranges, DHCP scopes, static reservations, and DNS with headroom for future VLANs, Pi-hole, NAS, lab servers, and VPN access. Use homelab-network-setup when escaping ISP-router-only setups, fixing double NAT, stabilizing Wi-Fi, or stopping servers from getting new addresses on every reboot. The skill produces a growth-ready foundation that pairs naturally with homelab-vlan-segmentation and homelab-pihole-dns follow-on work.6.5kinstalls 231Ios Icon Genios-icon-gen is an ECC community skill that builds PNG icon imagesets for Xcode asset catalogs from two sources: Apple SF Symbols (5000+ native symbols) and the Iconify API (275k+ open-source icons across 200+ collections). It searches collections, selects symbols, and outputs 1x, 2x, and 3x PNG imagesets ready for Assets.xcassets. iOS and macOS developers reach for ios-icon-gen when replacing placeholder tab bar or toolbar icons, prototyping UI chrome, or standardizing icon assets without hand-exporting every density from design tools.6.5kinstalls 232Agentic Osagentic-os is an ECC skill that treats Claude Code as a persistent runtime rather than a disposable chat session. It documents a four-layer architecture rooted in project directories: a kernel that routes tasks to specialist agents, slash-command interfaces, file-based JSON and Markdown memory, and scheduled automation—all without relying on an external database for state. Developers reach for agentic-os when building multi-agent workflows, personal or agentic operating systems, or long projects where context must survive session restarts. Trigger phrases include "agentic OS," "personal OS," "multi-agent," "agent coordinator," and "persistent agent." The pattern formalizes production agentic setups used inside ECC projects.6.5kinstalls 233Literature Reviewliterature-review is a skill from affaan-m/ecc for systematic academic and technical literature workflows spanning narrative, scoping, systematic, and meta-analysis review types. The workflow starts by turning prompts into searchable research questions, then screens sources, synthesizes findings, verifies citations, and logs evidence across peer-reviewed papers, preprints, patents, and technical reports. Developers default to scoping reviews for exploratory work and systematic reviews for publishable or clinical claims when rigor level is unspecified. Reach for literature-review when mapping state of the art, finding gaps, preparing cited background sections, or comparing contradictory evidence before greenlighting engineering investment.6.4kinstalls 234Ecc Guideecc-guide is the navigation skill for the Everything Claude Code ecosystem in the affaan-m/ecc repository. The skill explains how agents, skills, commands, hooks, rules, and install profiles relate, helps developers find the right ECC component for a task, and directs users to read live repository surfaces instead of relying on stale training data. Developers reach for ecc-guide when onboarding to ECC, choosing install profiles, asking what ECC includes, or resolving confusion about install paths and component boundaries. The skill is marked with community origin metadata and should be invoked before deep-diving into specialized ECC skills when the question is which tool to use.6.4kinstalls 235Ui To Vueui-to-vue is a community ECC skill that transforms directories of UI design screenshots and design exports into ready-to-use Vue 3 Composition API component code. Input is organized by module and page state, with supported asset folder names such as assets and icons for sliced images. The skill supports three component libraries: Vant for mobile-oriented UIs, Element Plus, and Ant Design Vue. Developers use ui-to-vue when migrating a design folder into initial page components, shared components, and starter router wiring for a Vue 3 codebase. It is built for batch conversion across many screens, not single-screenshot one-off components. The skill explicitly avoids projects that need deep interaction logic, data-flow design, accessibility review, or screenshots containing private customer data sent to external model APIs.6.4kinstalls 236Eval HarnessFormal evaluation framework for Claude Code sessions implementing eval-driven development EDD principles name eval-harness description Formal evaluation framework for Claude Code sessions implementing eval-driven development EDD principles allowed-tools Read Write Edit Bash Grep Glob Eval Harness Skill A formal evaluation framework for Claude Code sessions implementing eval-driven development EDD principles When to Activate Setting up eval-driven development EDD for AI-assisted workflows Defining pass fail criteria for Claude Code task completion Measuring agent reliability with pass k metrics Creating regression test suites for prompt or agent changes Benchmarking agent performance across model versions Philosophy Eval-Driven Development treats evals as the unit tests of AI development Define expected behavior BEFORE implementation Run evals continuously during development Track regressions with each change Use pass k metrics for reliability measurement Eval Types Capability Evals Test if Claude can do something it couldn't before markdown CAPABILITY EVAL feature-name Task Description of what Claude should accomplish Success Criteria Criterion 1 Criterion 2 Criterion 3 Expected O.6.4kinstalls 237Mle WorkflowMLE Workflow is an ECC agent skill for turning experimental notebook code into production-grade machine learning systems. It structures work around data contracts, reproducible training runs, measurable quality evaluation gates, deployable model artifacts, operational monitoring, and rollback procedures. Developers activate MLE Workflow when planning or reviewing production ML features such as ranking systems, recommenders, classifiers, embedding workflows, or model refresh cycles that must move beyond one-off Jupyter experiments. The skill uses Read, Write, Edit, Bash, Grep, and Glob to audit and harden ML codebases, ensuring training is repeatable, evaluation is gated, and deployed models remain observable in production.6.4kinstalls 238Skill Stocktakeskill-stocktake is an ECC slash command (/skill-stocktake) that audits Claude skills and commands for quality using a checklist plus AI holistic judgment. It offers Quick Scan for recently changed skills and Full Stocktake for a complete sequential subagent batch evaluation across global and project paths such as ~/.claude/skills/ and the current working directory. Developers reach for skill-stocktake when custom skills drift in quality, duplicate triggers, or lack clear activation criteria after rapid iteration. The command scopes paths relative to the invocation directory and targets both global and project-local skill trees. Results help teams retire weak skills, tighten descriptions, and align slash commands before wider agent rollout.6.4kinstalls 239Jpa PatternsJPA Hibernate patterns for entity design relationships query optimization transactions auditing indexing pagination and pooling in Spring Boot name jpa-patterns description JPA Hibernate patterns for entity design relationships query optimization transactions auditing indexing pagination and pooling in Spring Boot origin ECC JPA Hibernate Patterns Use for data modeling repositories and performance tuning in Spring Boot When to Activate Designing JPA entities and table mappings Defining relationships OneToMany ManyToOne ManyToMany Optimizing queries N 1 prevention fetch strategies projections Configuring transactions auditing or soft deletes Setting up pagination sorting or custom repository methods Tuning connection pooling HikariCP or second-level caching Entity Design java Entity Table name markets indexes Index name idx_markets_slug columnList slug unique true EntityListeners AuditingEntityListener class public class MarketEntity Id GeneratedValue strategy GenerationType IDENTITY private Long id Column nullable false length 200 private String name Column nullable false unique true length 120 private String slug Enumerated EnumType STRING private MarketStatus status MarketStatus.6.4kinstalls 240Network Interface Healthnetwork-interface-health is an ECC community skill for diagnosing physical-link and switch-port problems that masquerade as application or routing failures. The skill teaches baseline-then-delta counter analysis: capture interface statistics, wait a measurement interval, capture again, and compare increments rather than trusting absolute totals. It documents Cisco-style show interfaces commands, Linux ip -s link, ethtool, and ethtool -S queries, plus SNMP counters such as ifInErrors, ifOutErrors, and ifOutDiscards. A counter reference table explains eight error types—CRC, runts, giants, input drops, output drops, resets, and collisions—with common causes and remediation order. Diagnosis flows separate CRC and input-error investigation from congestion-driven output drops and duplex or speed negotiation mismatches. A bundled Python parser safely slices show interfaces blocks without arbitrary character windows. Developers reach for network-interface-health when monitoring reports rising interface errors or intermittent reachability on a VLAN or WAN link.6.4kinstalls 241Angular Developerangular-developer is an ECC skill that generates Angular code and provides architectural guidance across projects, components, services, directives, pipes, guards, and resolvers. It covers modern reactivity with Angular Signals, linkedSignal, and resource APIs, plus dependency injection, routing, SSR, accessibility with ARIA, animations, component styles, Tailwind CSS, testing, and Angular CLI tooling. Developers reach for angular-developer when scaffolding a new Angular app or library, implementing signal-based state, building typed forms, or aligning generated code with current Angular best practices. The skill triggers on any Angular codebase work so agents produce framework-correct patterns instead of generic React-style component code.6.4kinstalls 242Netmiko Ssh Automationnetmiko-ssh-automation is an ECC-origin agent skill for Python network automation with Netmiko. It guides Claude through SSH connection patterns, device-type selection, command execution, output parsing, and safe review of scripts targeting routers, switches, and similar gear. Developers reach for netmiko-ssh-automation when automating show commands, config pushes, or health checks across a fleet without hand-writing fragile Paramiko boilerplate. The skill emphasizes safe connection handling, credential hygiene, and script review before production runs. It fits network engineers and backend developers maintaining infrastructure-as-code or ops tooling in Python.6.4kinstalls 243Cisco Ios Patternscisco-ios-patterns is an ECC community skill that codifies Cisco IOS and IOS-XE review workflows for planned changes, troubleshooting, and incident response. It structures read-only show command selection, config hierarchy explanation (global, interface, routing process, line modes), ACL wildcard mask and interface direction checks, and maintenance-window validation before writes hit running config. The workflow prefers capturing current state, reviewing exact candidate configs, confirming out-of-band access, and verifying intentional saves—treating examples as patterns rather than copy-paste production configs. Network engineers reach for cisco-ios-patterns when reviewing router or switch snippets, building change checklists, or explaining how to collect evidence without worsening an outage.6.4kinstalls 244Plan Orchestrateplan-orchestrate is a planning skill that turns any multi-step plan document into ready-to-paste /orchestrate custom prompts without manually selecting agents for each step. Developers write or import a structured plan, and the skill formats orchestration prompts that dispatch the right agent work per stage. Reach for plan-orchestrate when a written implementation plan already exists and you want to batch-convert steps into Claude Code orchestration commands. The ECC repo notes some localized readme copies are still being translated, but the skill targets English plan-to-orchestrate prompt generation.6.4kinstalls 245Homelab Network Readinesshomelab-network-readiness is a community ECC planning skill for home and small-lab networks that mix VLANs, local DNS resolvers like Pi-hole or Unbound, firewall rules, and remote VPN access. The skill reviews current topology, rollback paths, console access, and maintenance windows before recommending VLAN splits across trusted, IoT, guest, server, or management segments. It deliberately stops short of copy-paste router or firewall configuration until platform details and rollback plans are confirmed. Developers reach for homelab-network-readiness when moving from a flat LAN to segmented VLANs or when migrating DHCP clients to a local DNS resolver. The workflow emphasizes security evaluation and performance testing as pre-change gates.6.4kinstalls 246Clickhouse IoClickHouse database patterns query optimization analytics and data engineering best practices for high-performance analytical workloads name clickhouse-io description ClickHouse database patterns query optimization analytics and data engineering best practices for high-performance analytical workloads ClickHouse 分析パターン高性能分析とデータエンジニアリングのためのClickHouse固有のパターン概要 ClickHouseはオンライン分析処理 OLAP 用のカラム指向データベース管理システム DBMS です大規模データセットに対する高速分析クエリに最適化されています主な機能カラム指向ストレージデータ圧縮並列クエリ実行分散クエリリアルタイム分析テーブル設計パターン MergeTreeエンジン最も一般的 sql CREATE TABLE markets_analytics date Date market_id String market_name String volume UInt64 trades UInt32 unique_traders UInt32 avg_trade_size Float64 created_at DateTime ENGINE MergeTree PARTITION BY toYYYYMM date ORDER BY date market_id SETTINGS index_granularity 8192 ReplacingMergeTree 重複排除 sql 重複がある可能性のあるデータ複数のソースからなど用 CREATE TABLE user_events event_id String user_id String event_type String timestamp DateTime properties String ENGINE ReplacingMergeTree PARTITION BY toYYYYMM timestamp ORDER BY user_id event_id timestamp PRIMARY KEY user_id event_id AggregatingMergeTree 事前集計 sql 集計メトリクスの維持用 CREATE TABLE market_stats_hourly hour DateTime market_id String tota.6.4kinstalls 247Homelab Wireguard Vpnhomelab-wireguard-vpn is a community ECC skill for deploying WireGuard as a fast, modern VPN for homelab remote access. The skill walks through server setup on Raspberry Pi, Linux hosts, pfSense, or routers, generating WireGuard keypairs, writing peer configuration files, and applying iptables forwarding rules with correct key file permissions. Developers reach for homelab-wireguard-vpn when they need simpler, faster remote access than OpenVPN for phones and laptops connecting to a home network. All examples are reviewed before application and changes are expected during a defined maintenance window. The workflow emphasizes end-to-end encryption and practical homelab access patterns.6.3kinstalls 248Network Config Validationnetwork-config-validation is an ECC skill that statically reviews router and switch configuration files before they reach production. It flags dangerous commands, duplicate IP assignments, overlapping subnets, and stale interface or neighbor references that commonly cause outages during cutovers. Network engineers and platform teams reach for network-config-validation when reviewing candidate configs from templates, automations, or peer review—especially when a bad push could black-hole traffic or duplicate addressing across VLANs. The published SKILL.md excerpt notes an in-progress Japanese translation, but the skill description and intended validation workflow remain the authoritative capability.6.3kinstalls 249Quarkus Patternsquarkus-patterns is an ECC skill documenting Quarkus 3.x LTS architecture for cloud-native, event-driven Java services. It covers JAX-RS and RESTEasy Reactive APIs structured as resource → service → repository layers, Apache Camel routes with RabbitMQ, Hibernate Panache persistence, caching, reactive streams, validation, exception mapping, and pagination. Developers activate it when scaffolding Quarkus microservices instead of assembling patterns from scattered docs. The skill README is Spanish-language but the patterns target standard Quarkus 3.x APIs familiar to Java backend engineers.6.3kinstalls 250Skill Scoutskill-scout is an affaan-m ECC community skill that developers invoke before building a new agent skill from scratch. It extracts the target task, trigger conditions, related frameworks, and three to five search keywords, then queries local skill folders, marketplaces, GitHub repositories, and the broader web for matches to review, adopt, or fork. The workflow prevents duplicate community skills while still requiring proper inspection before external adoption. Developers reach for skill-scout when they ask whether a skill already exists, propose a new workflow skill, or want to extend an existing package. If a user explicitly demands greenfield creation, the skill confirms that choice then proceeds with the requested build path instead of searching further.6.3kinstalls 251Cpp TestingGet an agent that writes, fixes, configures, and debugs GoogleTest-based C++ unit and integration tests following modern TDD practices.6.3kinstalls 252Cost Trackingcost-tracking is a Claude Code skill that analyzes Claude Code spending from the local SQLite database at ~/.claude-cost-tracker/usage.db. It assumes a cost-tracking hook or plugin already writes usage rows to that database, then answers questions about session cost, token usage, budgets, overages, and breakdowns by project, tool, session, model, or date. Developers can compare today versus yesterday, inspect recent trends, and export recent usage records to CSV. The skill originated from community PR #1304 by MayurBhavsar and uses sqlite3 to query stored metrics when users ask how much they spent.6.3kinstalls 253Network Bgp Diagnosticsnetwork-bgp-diagnostics is an affaan-m/ecc skill for safely diagnosing Border Gateway Protocol problems on routers and cloud network gateways. It helps operators inspect BGP neighbor state, advertised and received routes, and routing policy mismatches using non-disruptive diagnostic commands suited to production environments. Developers and SREs reach for network-bgp-diagnostics when hybrid cloud or on-prem BGP sessions flap, prefixes disappear, or asymmetric routing appears after policy changes. The skill emphasizes safe command selection so agents do not suggest clearing sessions or other destructive fixes during initial triage.6.3kinstalls 254Quarkus Securityquarkus-security is an ECC-origin skill from affaan-m/ecc for securing Quarkus applications with authentication, authorization, and input validation best practices. It guides JWT and OIDC setup, @RolesAllowed and SecurityIdentity RBAC, Bean Validation and custom validators, CORS and security headers, and secrets handling through Vault, environment variables, and configuration sources. Developers reach for quarkus-security when adding auth to Quarkus services or auditing existing endpoints before production deploy. The skill focuses on Java microservice appsec patterns rather than generic OWASP checklists without framework context.6.3kinstalls 255Flox Environmentsflox-environments teaches agents to provision reproducible dev environments using Flox, a Nix-based declarative environment manager. Use it when projects need system-level dependencies—compilers, OpenSSL, libvips, BLAS, LAPACK—or pinned toolchains for Python, Node.js, Rust, Go, C/C++, Java, Ruby, Elixir, PHP, and more across macOS and Linux. Flox lets agents install project-scoped tools without sudo, system pollution, or sandbox limits, then commit .flox/ and manifest.toml so any developer reproduces the environment with flox activate. It also covers co-locating local services like PostgreSQL, Redis, and Kafka beside dev tools to fix works-on-my-machine drift.6.3kinstalls 256Quarkus Tddquarkus-tdd is an ECC-origin agent skill for test-driven development on Quarkus 3.x LTS event-driven microservices. The workflow sequences JUnit 5 unit tests, Mockito doubles, REST Assured HTTP checks, Apache Camel route tests, and JaCoCo coverage gates targeting 80%+ combined unit and integration coverage. Reach for quarkus-tdd when adding REST endpoints, fixing bugs, refactoring reactive or Camel-backed services, or validating security and data-access logic in Quarkus projects. The skill is optimized for event-driven architectures where Camel routes and async handlers need the same TDD rigor as synchronous REST layers.6.3kinstalls 257Tinystruct Patternstinystruct-patterns is a development skill from the ECC harness for engineers building on the tinystruct Java framework—a lightweight runtime that treats CLI and HTTP as equal citizens with no main() boilerplate. The skill walks through creating Application modules extending AbstractApplication, mapping @Action endpoints for terminal and web, wiring ActionRegistry path injection, serializing JSON via Builder/Builders, persisting with AbstractData POJOs and XML mapping, and handling SSE, file uploads, events, and outbound HTTP. It bundles six reference guides covering architecture, routing, data handling, database persistence, system usage, and JUnit 5 unit plus HTTP integration testing. Reach for tinystruct-patterns when extending tinystruct itself or shipping hybrid CLI/web services that share one codebase. The skill does not replace tinystruct docs—it gives agents opinionated patterns aligned with zero-dependency JSON and dual-mode design.6.3kinstalls 258Homelab Vlan Segmentationhomelab-vlan-segmentation is a community ECC skill for developers running home labs or small offices who want VLAN-based isolation between IoT gadgets, guest Wi-Fi, trusted workstations, and servers. It explains how to add firewall rules that block lateral movement between segments while preserving existing protections, and it stresses applying changes in a maintenance window with connectivity checks after each step. The skill covers UniFi, pfSense, OPNsense, and MikroTik setups where smart bulbs, cameras, TVs, and lab servers should not share one broadcast domain. Reach for homelab-vlan-segmentation when flattening IoT and trusted devices on one subnet creates unnecessary attack surface.6.3kinstalls 259Quarkus Verificationquarkus-verification is an ECC-origin agent skill describing a Spanish-language verification loop for Quarkus microservice projects. Developers trigger it before opening pull requests, after major refactors or dependency upgrades, and before staging or production deploys. The pipeline order is build, lint static analysis, tests with coverage thresholds, security scanning, native compilation, and diff review against the target branch. The skill fits JVM teams shipping Quarkus REST or reactive services who want agents to run the same pre-PR checklist every time instead of skipping native builds or security scans under deadline pressure. It validates that coverage and compile targets pass before merge, reducing regressions on GraalVM native images.6.2kinstalls 260Homelab Pihole Dnshomelab-pihole-dns is a community ECC skill for developers running Pi-hole as network-wide DNS on a Raspberry Pi or any Linux host. Every device on the LAN inherits ad and malware domain blocking without per-browser extensions. The skill covers Pi-hole installation, pointing the home network DNS to Pi-hole, managing blocklists, configuring DNS-over-HTTPS upstream resolvers, and creating local records such as nas.home.lan or pi.home.lan. It also troubleshoots devices that lose internet access after Pi-hole deployment. Use homelab-pihole-dns after homelab-network-setup when you need centralized DNS privacy, ad blocking, and friendly local names for NAS and lab servers.6.2kinstalls 261Iterative Retrievalサブエージェントのコンテキスト問題を解決するためにコンテキスト取得を段階的に洗練するパターン name iterative-retrieval description サブエージェントのコンテキスト問題を解決するためにコンテキスト取得を段階的に洗練するパターン反復検索パターンマルチエージェントワークフローにおけるコンテキスト問題を解決しますサブエージェントは作業を開始するまでどのコンテキストが必要かわかりません問題サブエージェントは限定的なコンテキストで起動されます以下を知りませんどのファイルに関連するコードが含まれているかコードベースにどのようなパターンが存在するかプロジェクトがどのような用語を使用しているか標準的なアプローチは失敗しますすべてを送信コンテキスト制限を超える何も送信しないエージェントに重要な情報が不足必要なものを推測しばしば間違い解決策反復検索コンテキストを段階的に洗練する4フェーズのループ DISPATCH EVALUATE LOOP REFINE 最大3サイクルその後続行フェーズ1 DISPATCH 候補ファイルを収集する初期の広範なクエリ javascript 高レベルの意図から開始 const initialQuery patterns src ts lib ts keywords authentication user session excludes test ts spec ts 検索エージェントにディスパッチ const candidates await retrieveFiles initialQuery フェーズ2 EVALUATE 取得したコンテンツの関連性を評価 javascript function evaluateRelevance files task return files map file path file path relevance scoreRelevance file content task reason explainRelevance file content task missingContext identifyGaps file content task スコアリング基準高 0 8-1 0 ターゲット機能を直接実装中 0 5-0 7 関連するパターンや型を含む低 0 2-0 4 間接的に関連なし 0-0 2 関連なし除外フェーズ3 REFINE 評価に基づいて検索基準を更新 javascript function refineQuery evaluation previousQuery return 高関連性ファイルで発見された新しいパターンを追加 patterns previousQuery patterns ext.6.2kinstalls 262Fsharp Testingfsharp-testing is an ECC skill documenting comprehensive testing patterns for F# applications on the standard .NET test stack. It maps four core tools—xUnit as the test framework, FsUnit.xUnit for F#-friendly assertions, Unquote for quotation-based failure messages, and FsCheck.xUnit for property-based tests—plus modern .NET practices for integration coverage. Developers activate fsharp-testing when writing new F# tests, reviewing test quality, setting up test infrastructure, or debugging flaky and slow suites. The skill emphasizes functional approaches such as property tests alongside example-based unit and integration tests. Reach for fsharp-testing whenever F# services need reliable automated coverage rather than ad hoc C#-style assertion patterns.6.2kinstalls 263Verification LoopA comprehensive verification system for Claude Code sessions Invoke this skill: - After completing a feature or significant code change - Before creating a PR - When you want to ensure quality gates pass - After refactoring ## Verification Phases ### Phase 1: Build Verification ```bash # Check if project builds npm run build 2>&1 | tail -20 # OR pnpm build 2> A comprehensive verification system for Claude Code sessions. ## When to Use Invoke this skill: - After completing a feature or significant code change - Before creating a PR - When you want to ensure quality gates pass - After refactoring ## Verification Phases ### Phase 1: Build Verification ```ba # Verification Loop Skill A comprehensive verification system for Claude Code sessions. ## When to Use Invoke this skill: - After completing a feature or significant code change - Before creating a PR - When you want to ensure quality gates pass - After refactoring ## Verification Phases ### Phase 1: Build Verification ```bash # Check if project builds npm run build 2>&1 | tail -20 # OR pnpm build 2>&1 | tail -20 ``` If build fails, STOP and fix before continuing.6.2kinstalls 264Configure EccEverything Claude Code のインタラクティブなインストーラースキルとルールの選択とインストールをユーザーレベルまたはプロジェクトレベルのディレクトリへガイドしパスを検証し必要に応じてインストールされたファイルを最適化します name configure-ecc description Everything Claude Code のインタラクティブなインストーラースキルとルールの選択とインストールをユーザーレベルまたはプロジェクトレベルのディレクトリへガイドしパスを検証し必要に応じてインストールされたファイルを最適化します Configure Everything Claude Code ECC Everything Claude Code プロジェクトのインタラクティブなステップバイステップのインストールウィザードです AskUserQuestion を使用してスキルとルールの選択的インストールをユーザーにガイドし正確性を検証し最適化を提供します起動タイミングユーザーが configure ecc install ecc setup everything claude code などと言った場合ユーザーがこのプロジェクトからスキルまたはルールを選択的にインストールしたい場合ユーザーが既存の ECC インストールを検証または修正したい場合ユーザーがインストールされたスキルまたはルールをプロジェクト用に最適化したい場合前提条件このスキルは起動前に Claude Code からアクセス可能である必要がありますブートストラップには2つの方法があります 1 プラグイン経由 plugin install ecc ecc プラグインがこのスキルを自動的にロードします 2 手動このスキルのみを claude skills configure-ecc SKILL md にコピーし configure ecc と言って起動しますステップ 0 ECC リポジトリのクローンインストールの前に最新の ECC ソースを tmp にクローンします bash rm rf tmp everything-claude-code git clone https github com affaan-m everything-claude-code git tmp everything-claude-code 以降のすべてのコピー操作のソースとして ECC_ROOT tmp everything-claude-code を設定しますクローンが失敗した場合ネットワークの問題など AskUserQuestion を使用してユーザーに既存の ECC クローンへのローカルパスを提供するよう依頼しますステップ 1 インストールレベルの選択.6.2kinstalls 265Search FirstEnsure they discover existing tools, libraries, and patterns before writing new code from scratch.6.2kinstalls 266Prisma Patternsprisma-patterns is an ECC skill with production Prisma ORM guidance for TypeScript backends, tested against Prisma 5.x and 6.x with noted differences from Prisma 4. It documents schema design, query optimization, transactions, pagination, and non-obvious traps: `updateMany` returning a count not records, `$transaction` timeouts, `migrate dev` resetting the database, `@updatedAt` skipped on bulk writes, and serverless connection exhaustion. Prisma 5's `relationJoins` preview can load related records in fewer queries. Developers check versions with `npx prisma --version` before applying version-specific advice. Reach for prisma-patterns when agents generate Prisma code and you want battle-tested guardrails instead of rediscovering ORM footguns.6.2kinstalls 267Rust TestingGenerate comprehensive Rust tests following TDD with unit, integration, async, property-based, and mocked tests.6.2kinstalls 268Nutrient Document ProcessingNutrient DWS API を使用してドキュメントの処理変換 OCR 抽出編集署名フォーム入力を行います PDF DOCX XLSX PPTX HTML 画像に対応しています name nutrient-document-processing description Nutrient DWS API を使用してドキュメントの処理変換 OCR 抽出編集署名フォーム入力を行います PDF DOCX XLSX PPTX HTML 画像に対応しています Nutrient Document Processing Nutrient DWS Processor API https www nutrient io api でドキュメントを処理しますフォーマット変換テキストとテーブルの抽出スキャンされたドキュメントの OCR PII の編集ウォーターマークの追加デジタル署名 PDF フォームの入力が可能ですセットアップ nutrient io https dashboard nutrient io sign_up product processor で無料の API キーを取得してください bash export NUTRIENT_API_KEY pdf_live_ すべてのリクエストは https api nutrient io build に instructions JSON フィールドを含むマルチパート POST として送信されます操作ドキュメントの変換 bash DOCX から PDF へ curl X POST https api nutrient io build H Authorization Bearer NUTRIENT_API_KEY F document docx document docx F instructions parts file document docx o output pdf PDF から DOCX へ curl X POST https api nutrient io build H Authorization Bearer NUTRIENT_API_KEY F document pdf document pdf F instructions parts file document pdf output type docx o output docx HTML から PDF へ curl X POST https api nutrient io build H Authorization Bearer NUTRIENT_API_KEY F6.2kinstalls 269Codebase OnboardingQuickly understand an unfamiliar repository and generate a ready-to-use CLAUDE.md that lets Claude Code operate effectively from day one.6.1kinstalls 270Django VerificationVerification loop for Django projects migrations linting tests with coverage security scans and deployment readiness checks before release or PR name django-verification description Verification loop for Django projects migrations linting tests with coverage security scans and deployment readiness checks before release or PR Django 検証ループ PR前大きな変更後デプロイ前に実行して Djangoアプリケーションの品質とセキュリティを確保しますフェーズ1 環境チェック bash Pythonバージョンを確認 python version プロジェクト要件と一致すること仮想環境をチェック which python pip list outdated 環境変数を確認 python c import os import environ print DJANGO_SECRET_KEY set if os environ get DJANGO_SECRET_KEY else MISSING DJANGO_SECRET_KEY 環境が誤って構成されている場合は停止して修正しますフェーズ2 コード品質とフォーマット bash 型チェック mypy config-file pyproject toml ruffでリンティング ruff check 自動修正 Django固有のチェック python manage py check deploy 一般的な問題パブリック関数の型ヒントの欠落 PEP 8フォーマット違反ソートされていないインポート本番構成に残されたデバッグ設定フェーズ3 マイグレーション bash 未適用のマイグレーションをチェック python manage py showmigrations 欠落しているマイグレーションを作成 python manage py makemigrations check マイグレーション適用のドライラン python manage py migrate plan マイグレーションを適用テスト環境 python manage py migrate マイグレーションの競合をチェック python manage py makemigrations merge 競合がある場合のみレポート保留中のマイグレーション数マイグレーションの競合マイグレーションのないモデルの変更フェーズ4.6.1kinstalls 271Agent Harness ConstructionAI エージェントのアクション空間ツール定義観測フォーマットを設計最適化して完了率を向上させます name agent-harness-construction description AI エージェントのアクション空間ツール定義観測フォーマットを設計最適化して完了率を向上させます origin ECC エージェントハーネス構築エージェントの計画ツール呼び出しエラーからの回復完了への収束を改善する場合にこのスキルを使用しますコアモデルエージェントの出力品質は以下によって制約されます 1 コンテキストバジェットの品質アクション空間の設計 1 分離が不可能な場合を除きキャッチオールツールは避ける粒度ルール高リスク操作デプロイマイグレーション権限にはマイクロツールを使用する一般的な編集読み取り検索ループには中規模ツールを使用するラウンドトリップのオーバーヘッドが支配的なコストである場合のみマクロツールを使用する観測の設計すべてのツールレスポンスに含めるべき内容 status success warning error summary 一行の結果 next_actions 実行可能なフォローアップ artifacts ファイルパス ID エラー回復コントラクトすべてのエラーパスに含めるべき内容根本原因のヒント安全なリトライ指示明示的な停止条件コンテキストバジェット管理 1 大きなガイダンスはオンデマンドで読み込まれるスキルに移動する 3 長いドキュメントをインラインで挿入するよりファイルへの参照を優先する 4 任意のトークン閾値ではなくフェーズの境界でコンパクト化するアーキテクチャパターンガイダンス ReAct 不確実なパスを持つ探索的タスクに最適関数呼び出し構造化された決定論的フローに最適ハイブリッド推奨 ReAct 計画型付きツール実行ベンチマーク追跡すべき指標完了率タスクあたりのリトライ数 pass 1 および pass 3 成功タスクあたりのコストアンチパターンセマンティクスが重複するツールが多すぎる回復ヒントのない不透明なツール出力次のステップなしのエラーのみの出力無関係な参照でコンテキストを過負荷にする name agent-harness-construction description AI エージェントのアクション空間ツール定義観測フォーマットを設計最適化して完了率を向上させます origin ECC エージェントハーネス構築エージェントの計画ツール呼び出しエラーからの回復完了への収束を改善する場合にこのスキルを使用しますコアモデルエージェントの出力品質は以下によって制約されます 1 コンテキストバジェットの品質アクション空間の設計.6.1kinstalls 272Agentic Engineeringname agentic-engineering description Operate as an agentic engineer using eval-first execution decomposition and cost-aware model routing Use when AI agents perform most implementation work and humans enforce quality and risk controls metadata origin ECC Agentic Engineering Use this skill for engineering workflows where AI agents perform most implementation work and humans enforce quality and risk controls Define completion criteria before execution Decompose work into agent-sized units Route model tiers by task complexity Measure with evals and regression checks Define capability eval and regression eval Run baseline and capture failure signatures Re-run evals and compare deltas Write test that captures desired behavior eval 2 Run test capture baseline failures 3 name agentic-engineering description Operate as an agentic engineer using eval-first execution decomposition and cost-aware model routing Use when AI agents perform most implementation work and humans enforce quality and risk controls metadata origin ECC Agentic Engineering Use this skill for engineering workflows where AI agents perform most implementation work and humans enforce quality and risk controls Define complet.6.1kinstalls 273Springboot VerificationRun a repeatable verification loop that builds, statically analyzes, tests with coverage, scans for security issues, and reviews diffs on Spring Boot services.6kinstalls 274Cost Aware Llm PipelineRoute LLM calls intelligently, track spend in real time, and keep API costs low while preserving output quality.6kinstalls 275Kotlin Coroutines FlowsThis skill teaches Kotlin coroutines and Flow patterns for Android and Kotlin Multiplatform projects. It explains structured concurrency with viewModelScope, coroutineScope plus async for parallel repository loads, and supervisorScope when one child failure must not cancel siblings. Flow coverage spans cold flows, StateFlow with SharingStarted.WhileSubscribed, combining multiple repositories into one UI state, and operators for debounce, distinctUntilChanged, flatMapLatest, catch, and exponential retryWhen. SharedFlow carries one-time snackbar and navigation effects to Compose collectors. Dispatcher guidance maps CPU work to Default, blocking IO to IO on JVM and Android only, and UI updates to Main, with KMP notes where IO is unavailable. Cancellation sections require ensureActive in long loops, finally cleanup for loading flags, and never swallowing CancellationException. Testing recipes use Turbine awaitItem assertions, runTest with advanceUntilIdle, and fake repositories backed by MutableStateFlow. Documented anti-patterns reject GlobalScope, init-block collection without scope, mutating lists inside MutableStateFlow, and creating flows in Composables without remember.6kinstalls 276React Performancereact-performance from affaan-m/ecc adapts Vercel Engineering React Best Practices (MIT, v1.0.0 from vercel-labs/agent-skills) into an agent skill for React 18/19 and Next.js performance work. The skill organizes 70+ rules across 8 priority categories: waterfalls, bundle size, server-side, client fetching, re-render, rendering, JS micro-perf, and advanced patterns. Developers use react-performance when writing, reviewing, or refactoring React and Next.js code for performance—not for generic CSS layout or backend API design. Rules cover eliminating request waterfalls, shrinking client bundles, optimizing server components, reducing unnecessary re-renders, and JS micro-optimizations aligned with Vercel engineering guidance. With 749 catalog installs, frontend engineers reach for react-performance during perf-sensitive refactors, PR reviews, and Next.js App Router tuning instead of ad hoc optimization guesses.6kinstalls 277Deep ResearchGenerate thorough, multi-source cited research reports on any topic using Firecrawl and Exa MCP tools.6kinstalls 278React Patternsreact-patterns is an ECC-origin skill for writing and reviewing robust React 18/19 applications. It covers hooks discipline, explicit Server versus Client component boundaries, Suspense paired with error boundaries, form actions, data-fetching choices, state-management decision trees, and accessibility-first composition. Developers activate it when editing function components, custom hooks, JSX or TSX files, migrating class components, or replacing heavy useEffect and forwardRef patterns. The skill helps choose between local state, context, and external stores with clear composition rules. Teams reach for react-patterns during feature work and code review when components must stay performant, accessible, and aligned with modern React conventions across React 18 and React 19 APIs.6kinstalls 279Autonomous LoopsSelect and implement the right autonomous Claude Code loop architecture for their project.5.9kinstalls 280Recsys Pipeline Architectrecsys-pipeline-architect in affaan-m/ecc is a spec-and-scaffold skill for composable recommendation, ranking, and feed pipelines encoding a six-stage pattern: Source, Hydrator, Filter, Scorer, Selector, and SideEffect. The framework generalizes beyond social feeds to any system picking the top K items for a user-context pair—content CMS ranking, RAG rerankers, task prioritizers, notification triage, search reranking, and ad ranking. Each stage has a defined responsibility: sourcing candidates, hydrating features, filtering ineligible items, scoring relevance, selecting the final set, and executing side effects like logging or writes. The pattern draws from xAI's open-sourced For You algorithm structure. Developers reach for recsys-pipeline-architect when greenfield ranking services need clear stage boundaries instead of monolithic recommenders. Output includes pipeline specs and scaffold code organized by stage interfaces. Triggers include feed ranking architecture, recommendation pipeline design, top-K selector patterns, and modular reranker scaffolding.5.9kinstalls 281React Testingreact-testing is an ECC (affaan-m/ecc) agent skill for behavior-focused React testing with React Testing Library, Vitest or Jest, MSW network mocking, and axe accessibility assertions. It defines decision boundaries between component tests and Playwright or Cypress end-to-end runs, helping developers test custom hooks, legacy untested components, and pages migrated from Enzyme patterns. The skill activates on requests to write tests, fix failing specs, add coverage, or review Vitest and RTL conventions. Developers reach for react-testing when building React SaaS frontends that need user-centric assertions, mocked API layers, and accessible UI validation before release.5.9kinstalls 282Frontend A11yfrontend-a11y is a community-origin Claude Code skill from affaan-m/ecc delivering practical accessibility patterns for React and Next.js. It covers semantic HTML, ARIA attributes, form labeling for input, select, and textarea elements, keyboard navigation, focus management, and screen reader support for modals, dropdowns, and interactive UI. The skill targets issues commonly flagged in code review: missing labels, incorrect ARIA, non-semantic interactive elements, and broken keyboard flows. Developers reach for frontend-a11y when building or reviewing forms, modals, dropdowns, or any interactive React and Next.js component requiring WCAG-aligned patterns.5.9kinstalls 283Laravel TddThe laravel-tdd skill teaches test-driven development for Laravel applications using PHPUnit and Pest with an 80 percent plus coverage target across unit, feature, and integration layers. The red-green-refactor cycle starts with a failing test, implements the minimal passing code, then refactors while keeping tests green. Unit tests cover pure PHP classes, value objects, and services. Feature tests target HTTP endpoints, authentication, validation, policies, and response shape. Integration tests combine database, queues, and external boundaries. Database strategy defaults to RefreshDatabase for feature and integration work, with DatabaseTransactions when schema is already migrated and DatabaseMigrations when full fresh runs are required. Prefer Pest for new tests unless the project standardizes PHPUnit. Factories with states supply test data. Fakes isolate jobs, queues, mail, notifications, and events. Http::fake handles external APIs. Sanctum actingAs covers API auth. Inertia responses use assertInertia helpers. Coverage runs via php artisan test with pcov or XDEBUG_MODE=coverage in CI.5.9kinstalls 284Swiftui PatternsThis skill documents modern SwiftUI architecture patterns for Apple platforms using the Observation framework, view composition, type-safe navigation, and performance tuning. Activation covers building views with @State, @Observable, and @Binding, designing NavigationStack flows, structuring view models and data flow, optimizing list rendering, and injecting dependencies via environment values. State management maps each property wrapper to its role, favoring @Observable over ObservableObject for property-level invalidation and @Environment over @EnvironmentObject for shared services. View composition splits screens into focused subviews and reusable ViewModifier styles to limit rerenders. NavigationStack with NavigationPath and a hashable Destination enum enables programmatic type-safe routing through an @Observable Router. Performance guidance uses LazyVStack, stable ForEach identifiers, .task for async work, Equatable views for expensive bodies, and #Preview macros with mock repositories. Documented anti-patterns reject legacy ObservableObject wrappers, async in body or init, child-owned view models, AnyView erasure, and ignored Sendable rules with actor data.5.8kinstalls 285Ai First EngineeringThis skill defines an AI-first engineering operating model for teams that release software with high-volume agent-generated implementation output. Process shifts elevate planning quality over typing speed, evaluation coverage over personal confidence, and review focus from syntax toward system behavior. Architecture guidance prioritizes agent-friendly systems with explicit boundaries, stable contracts, typed interfaces, and deterministic tests while avoiding implicit behavior spread through hidden conventions. Code review targets behavioral regression, security assumptions, data integrity, failure handling, and rollout safety while minimizing time on style issues already covered by automation. Strong engineers decompose ambiguous work, define measurable acceptance criteria, produce high-signal prompts and evaluations, and enforce risk controls under deadline pressure. Testing standards raise the bar for generated code with mandatory regression coverage on touched domains, explicit edge-case assertions, and integration checks at interface boundaries.5.8kinstalls 286Content EngineThe content-engine skill builds platform-native content systems for X, LinkedIn, TikTok, YouTube, newsletters, and repurposed multi-platform campaigns without flattening the author's real voice into platform slop. It starts from source material such as articles, notes, demos, docs, changelogs, transcripts, screenshots, or prior posts rather than generic post formulas. A source-first workflow extracts three to seven atomic claims, ranks them by sharpness and proof, assigns one strong idea per output, and adapts structure per platform while stripping platform-shaped filler. Platform rules compress X threads around advancing arguments, expand LinkedIn only enough for outsiders to follow, script short video around visual proof, and open newsletters with the point instead of warmup paragraphs. Voice handling defers to brand-voice when multiple outputs or reputation-sensitive launch copy need consistency. Hard bans delete hype phrases, fake engagement bait, forced casualness, and praise-stacking unless the user explicitly requests engagement tactics. Campaign deliverables include voice profiles when needed, core angle, platform-native drafts, optional posting order, and gaps before publ.5.8kinstalls 287Content Hash Cache PatternThis skill defines a content-hash file cache for expensive processing such as PDF parsing, text extraction, or image analysis. Cache keys are SHA-256 digests of file bytes, not paths, so renames and moves still hit cache while content edits invalidate automatically without an index file. Implementation chunks large files in 64KB reads, stores each entry as a frozen dataclass serialized to {hash}.json for O(1) lookup, and treats JSON corruption as a cache miss. A service-layer wrapper extract_with_cache checks the hash, returns cached documents on hit, otherwise calls a pure extract_text function and writes a new entry, preserving single responsibility. CLI tools can expose --cache and --no-cache by toggling cache_enabled. Best practices log truncated hashes on hit or miss, keep processors ignorant of caching, and never crash on bad cache files. Anti-patterns reject path-keyed dictionaries, embedding cache branches inside extraction functions, and dataclasses.asdict for nested frozen types. Skip when results must always be fresh, cache blobs would be huge, or outputs depend on parameters beyond raw file content.5.8kinstalls 288Regex Vs Llm Structured TextThis skill provides a practical decision framework for parsing structured text such as quizzes, forms, invoices, and documents by choosing between regular expressions and large language models. The core insight is that regex handles 95 to 98 percent of cases deterministically at low cost, reserving expensive LLM calls for low-confidence edge cases only. The documented architecture runs a regex parser first, applies confidence scoring to flag items below a 0.95 threshold, and routes only flagged items to a cheap LLM validator such as Claude Haiku. Production metrics from a 410-item quiz pipeline show 98 percent regex success with about five LLM calls and roughly 95 percent cost savings versus full-LLM parsing. Best practices mandate starting with regex even if imperfect, using TDD for parsers, returning new instances from cleaning steps, and recording regex success rate plus LLM call counts. Anti-patterns include sending all text to LLM when regex already covers 95 percent, using regex on highly variable free-form text, and skipping confidence scoring.5.8kinstalls 289Unclouduncloud is an ECC-origin agent skill referencing the uc CLI for decentralized self-hosting on personal clusters. The workflow covers bootstrapping machines with `uc machine`, deploying services from Compose files via `uc deploy`, publishing HTTP, HTTPS, and TCP ports, configuring Caddy reverse proxy ingress, adding static proxy routes for non-cluster devices, scaling workloads, inspecting logs, and managing volumes across WireGuard mesh-linked nodes. Reach for uncloud when operating Uncloud clusters through natural language instead of memorizing uc subcommands. The platform combines Docker containers, WireGuard mesh networking, and Caddy ingress for self-hosted SaaS and API deployments.5.8kinstalls 290Kotlin PatternsThe kotlin-patterns skill teaches idiomatic Kotlin for robust JVM and mobile codebases across seven areas: null safety with nullable types and safe calls, immutability via val and data class copy, sealed classes for exhaustive hierarchies, structured concurrency with coroutines and Flow, extension functions, type-safe DSLs with DslMarker, and Gradle Kotlin DSL build files. Activate when writing new Kotlin, reviewing pull requests, refactoring modules, designing libraries, or configuring build.gradle.kts. Guidance contrasts good and bad patterns for Elvis defaults versus force unwrap, expression bodies, value classes, scope functions, delegation, sequences, and collection pipelines. Coroutine sections document coroutineScope async parallelism, supervisorScope for independent child failures, Flow debounce and catch operators, and cancellation with ensureActive plus NonCancellable cleanup. Error handling favors sealed Result types, runCatching, require, and check over exceptions as control flow. Gradle examples wire Ktor, Exposed, Kotest, mockk, detekt, and JVM toolchain 21. Anti-patterns reject GlobalScope, nested lets, mutable data classes, and platform null leakage from Java.5.8kinstalls 291Rust PatternsThe rust-patterns skill teaches idiomatic Rust conventions for safe, performant, maintainable applications across ownership, errors, types, traits, concurrency, and crate layout. Ownership guidance favors borrowing references over unnecessary clones and uses Cow for flexible string ownership. Error handling mandates Result and question-mark propagation, thiserror for library errors, anyhow for applications, and forbids unwrap in production. Enums model valid states with exhaustive matching instead of wildcard catch-alls that hide new variants. Traits and generics accept generic inputs but return concrete types; trait objects suit plugin routers while monomorphization suits hot paths. Concurrency covers Arc Mutex for shared state, bounded mpsc channels, and Tokio async with timeout-aware HTTP fetches. Module organization groups code by domain auth, orders, db rather than by type, exposing minimal pub surfaces with pub crate for internals. Tooling references cargo check, clippy, fmt, test, audit, and bench workflows. Anti-patterns call out clone-to-satisfy-borrow-checker, blocking inside async, and Box dyn Error in libraries. Use when writing, reviewing, or refactoring Rust code or.5.8kinstalls 292Scholar Evaluationscholar-evaluation is an ECC community skill that scores academic and technical artifacts with structured 1-to-5 rubrics across methodology, evidence, citations, and limitations. It supports empirical papers, theoretical work, technical reports, systematic or narrative literature reviews, proposals, thesis chapters, and conference abstracts—with comprehensive, targeted, or comparative evaluation modes. Developers and researchers reach for scholar-evaluation when reviewing agent-generated research summaries, validating that claims are citation-backed, comparing multiple papers on the same rubric, or producing structured revision feedback. Each dimension scores from 1 (substantial revision needed) through 5 (publication-ready), making quality gaps explicit before outputs inform product or publication decisions.5.8kinstalls 293Marketing CampaignMarketing Campaign is an ECC-origin Claude skill for end-to-end launch planning and execution from one product brief. The skill guides audience research, positioning, campaign angle definition, landing page copy, email sequences, social posts, ad copy, short-form video scripts, and a content calendar as an orchestration layer across channels. Developers reach for Marketing Campaign when planning a product or feature launch, defining positioning before writing copy, or building a full content suite from a single brief. The workflow emphasizes campaigns that convert—not just campaigns that ship—by keeping messaging consistent across every touchpoint.5.8kinstalls 294Investor MaterialsInvestor-materials guides agents through creating and updating pitch decks, one-pagers, investor memos, accelerator applications, financial models, and fundraising materials that stay internally consistent across every asset. The skill enforces a single source of truth for traction metrics, pricing assumptions, raise size, use of funds, team bios, and milestones before any drafting begins. Its core workflow inventories canonical facts, surfaces missing assumptions, selects the asset type, drafts with explicit logic, and cross-checks every number against that truth. Pitch deck flow spans company wedge through problem, solution, product demo, market, business model, traction, team, competition, ask, and use of funds. Financial models require bear, base, and bull cases, layer-by-layer revenue logic, milestone-linked spending, and sensitivity analysis. Accelerator applications demand exact answers with traction and team advantage, avoiding puffery. A quality gate blocks delivery when numbers disagree, assumptions stay hidden, or claims become unverifiable.5.7kinstalls 295Pubmed Databasepubmed-database is a community skill from affaan-m/ecc that teaches coding agents to search MEDLINE through PubMed and NCBI E-utilities with reproducible Boolean queries. The workflow starts from a research question, splits concepts, and combines synonyms with AND/OR/NOT while applying field tags such as [ti], [ab], [tiab], [au], [ta], and MeSH filters for date or publication type. Developers reach for pubmed-database when they need PMID abstracts, citation metadata, literature monitoring, or systematic-review search paths callable from Python, shell, or any HTTP client. The skill replaces noisy general web search with API-backed biomedical retrieval suitable for life-science tooling, clinical software research, and evidence-backed engineering decisions.5.7kinstalls 296Continuous Agent LoopThe continuous-agent-loop skill is the v1.8+ standard pattern for sustained autonomous agent loops with quality gates, evaluation, and recovery controls, replacing autonomous-loops for one release while preserving compatibility. A loop selection flow routes strict CI and pull request work to continuous-pr, RFC decomposition to rfc-dag, exploratory parallel generation to infinite, and everything else to sequential as the default. Recommended production stacks chain four layers: RFC decomposition via ralphinho-rfc-pipeline, quality gates with plankton-code-quality plus /quality-gate, evaluation loops through eval-harness, and session persistence with nanoclaw-repl. Documented failure modes include loop churn without measurable progress, repeated retries on the same root cause, stalled merge queues, and cost drift from unbounded escalation. Recovery playbooks freeze the loop, run /harness-audit, shrink scope to the failing unit, and replay with explicit acceptance criteria before resuming autonomous work.5.7kinstalls 297Swift Actor PersistenceThis skill teaches patterns for thread-safe data persistence in Swift 5.9+ using actors, combining an in-memory cache with file-backed JSON storage. The core pattern is a generic LocalRepository actor keyed by Identifiable IDs: synchronous load on init, O(1) reads from cache, and atomic file writes on save or delete. All public calls are async due to actor isolation, which replaces manual locks or DispatchQueue synchronization with compiler-enforced safety. The docs include a full actor implementation, usage examples, and integration with @Observable ViewModels for reactive UI updates. Design decisions cover Sendable types, minimal public APIs, atomic writes to prevent corruption, and anti-patterns like nonisolated bypasses. Activate when building offline-first iOS or macOS apps, local user data stores, or replacing legacy concurrency code with modern Swift actor-based repositories. The table of design decisions explains why actors beat class-plus-lock approaches and why dictionary keyed lookups keep reads fast while durable JSON on disk survives restarts.5.7kinstalls 298Swift Protocol Di TestingProtocol-based dependency injection for testable Swift code mock file system network and external APIs using focused protocols and Swift Testing Any Swift code that touches file system network or external APIs Testing error handling paths that are hard to trigger in real environments Building modules that need to work in app test and SwiftUI preview contexts Apps using Swift concurrency actors structured concurrency that need Patterns for making Swift code testable by abstracting external dependencies file system network iCloud behind small focused protocols Enables deterministic tests without I O When to Activate Writing Swift code that accesses file system network or external APIs Need to test error han Swift Protocol-Based Dependency Injection for Testing Patterns for making Swift code testable by abstracting external dependencies file system network iCloud behind small focused protocols Enables deterministic tests without I O When to Activate Writing Swift code that accesses file system network or external APIs Need to test error handling paths without triggering real failures Building modules that work across environments app test5.7kinstalls 299Compose Multiplatform PatternsThe compose-multiplatform-patterns skill teaches Jetpack Compose and Compose Multiplatform patterns for shared UI across Android, iOS, desktop, and web. State management uses a ViewModel with a single screen state data class exposed as StateFlow, collected in Compose via collectAsStateWithLifecycle, plus a sealed event interface replacing many callback lambdas. Navigation covers type-safe Serializable routes in Compose Navigation 2.8+, dialog destinations for confirmations, and passing navigation lambdas instead of NavController into deep composables. Composable design favors slot-based APIs and strict Modifier ordering from layout padding through shape, drawing, and interaction. KMP platform UI uses expect and actual composables for status bar styling. Performance guidance marks stable types with @Immutable, assigns LazyColumn item keys, defers reads with derivedStateOf, and uses remember to avoid recomposition allocations. Theming applies Material 3 dynamic color on Android. Documented anti-patterns reject mutableStateOf in ViewModel when StateFlow exists, heavy work inside composables, and LaunchedEffect(Unit) substituting ViewModel init. References android-clean-architecture f.5.7kinstalls 300Investor OutreachThe investor-outreach skill drafts short, concrete investor-facing messages for fundraising: cold emails, warm intro blurbs, follow-ups, update emails, and post-meeting notes. Activation covers writing to angels, VCs, strategic investors, or accelerators when the user needs personalized outbound copy with a clear ask. Core rules require personalization on every message, low-friction asks, proof over adjectives, tight word count, and copy that could not be sent unchanged to any investor. Hard bans flag phrases like generic thesis praise, vague founder adjectives, begging language, and soft closing questions when a direct ask works better. Cold emails follow a five-part structure: specific subject, why this investor, company pitch with proof, one concrete next step, and a credibility sign-off. Personalization must cite portfolio companies, public thesis, mutual connections, or product-market fit; if context is missing the skill says so. Default follow-up cadence is day zero outbound, day four or five with a new data point, and day ten to twelve with a clean close. Warm intro requests include a forwardable blurb under one hundred words.5.6kinstalls 301Nextjs TurbopackThe nextjs-turbopack skill documents Next.js 16+ development with Turbopack, the Rust incremental bundler enabled by default for local dev. It explains when to stay on Turbopack versus fall back to webpack with --webpack or --no-turbopack when hitting Turbopack bugs or webpack-only plugins. File-system caching under .next reuses prior work so restarts run five to fourteen times faster on large projects. Covers next dev, next build, and next start commands plus experimental Bundle Analyzer in Next.js 16.1+ for inspecting heavy dependencies and code-splitting. Best practices recommend recent 16.x releases, avoiding unnecessary cache clears, App Router with server components, and official bundle analysis for production size issues. Use when developing or debugging Next.js 16+ apps, diagnosing slow dev startup or HMR, or optimizing production bundles. Production build tooling varies by Next.js version; check official docs for Turbopack versus webpack behavior on next build.5.6kinstalls 302Foundation Models On DeviceThe foundation-models-on-device skill documents patterns for integrating Apple's on-device language model into iOS apps using the FoundationModels framework on Apple Intelligence. It covers checking SystemLanguageModel availability before sessions, creating LanguageModelSession for single-turn and multi-turn flows with role instructions, structured generation via @Generable types with @Guide range and count constraints, custom Tool calling for domain-specific actions, and snapshot streaming with PartiallyGenerated types for real-time SwiftUI updates. Key design notes include on-device execution for privacy and offline use, a 4096 token limit requiring chunked inputs, one request per session via isResponding, and accessing results through response.content. Best practices stress availability checks, instruction tuning, GenerationOptions temperature, and Instruments profiling. Anti-patterns include concurrent session requests, raw string parsing when @Generable applies, and assuming model readiness across devices. Activate for privacy-sensitive text generation, structured extraction from natural language, offline AI features, progressive streaming UI, and tool-augmented domain lookup.5.6kinstalls 303Bun RuntimeBun Runtime is a Claude Code agent skill for Bun as an all-in-one JavaScript toolkit: runtime, package manager, bundler, and test runner. Use it when adopting Bun, migrating from Node, writing or debugging Bun scripts and tests, or configuring Bun on Vercel. The skill contrasts when to prefer Bun for new JS/TS projects, fast install and run speed, a single run-install-test-build toolchain, and Vercel Bun runtime support versus Node for maximum ecosystem compatibility, legacy tooling, or known Bun dependency issues. Migration guidance replaces node with bun run, npm install with bun install, npm scripts with bun run, and npx with bun x. Documented workflows cover bun install lockfiles (bun.lock or bun.lockb), env-file script runs, bun test with a Jest-like API, Bun.file and Bun.serve runtime APIs, and Vercel settings using bun install --frozen-lockfile plus bun run build or bun build output. Best practices commit lockfiles, prefer bun run for scripts, run TypeScript natively, and keep dependencies current as Bun evolves.5.6kinstalls 304Swift Concurrency 6 2Swift 6.2 approachable concurrency runs code single-threaded by default and treats parallel execution as an explicit opt-in through @concurrent. Async functions stay on the caller actor instead of implicitly offloading to background threads, which fixes data-race errors that looked safe in Swift 6.1 MainActor models such as calling photoProcessor from an async MainActor method. The skill documents isolated protocol conformance so MainActor types adopt protocols without unsafe workarounds, MainActor default inference for app targets, and @MainActor protection for global and static mutable state. Enable SE-0466 and SE-0461 approachable concurrency build settings in Xcode 26 or SPM before adopting these patterns; without them, sample code still contains real data races the compiler flags. Use @concurrent only for CPU-heavy paths after Instruments profiling. Migration uses swift.org tools, incremental feature toggles, and compile-time data-race errors as guardrails. Start MainActor-centric for typical UI apps, then offload identified hot paths.5.6kinstalls 305Architecture Decision RecordsThe architecture-decision-records skill records architecture choices as structured ADR documents alongside code instead of leaving rationale in Slack threads, PR comments, or memory. It activates when users ask to record a decision, pick among frameworks, libraries, databases, or API designs, explain why X over Y, or review why a past choice was made. New ADRs follow a Michael Nygard template adapted for AI-assisted development with Context, Decision, Alternatives Considered, and Consequences including risks. The workflow initializes docs/adr/ only with user consent, extracts the core choice, documents rejected options, assigns incremental numbers, presents a draft for explicit approval before writing docs/adr/NNNN-decision-title.md, and updates README.md index. Reading mode scans the index and surfaces Context and Decision sections or offers to create a missing record. Decision signals include explicit record requests and implicit framework or auth strategy comparisons. Good ADRs stay specific, cite rationale, list rejected alternatives, note tradeoffs, and remain readable in about two minutes. Lifecycle states are proposed, accepted, deprecated, or superseded.5.6kinstalls 306Visa Doc TranslateThe visa-doc-translate skill converts visa application document images into English translations packaged as a bilingual PDF. When given an image path it runs automatically without per-step confirmation. HEIC files convert to PNG via sips; EXIF orientation is read and images rotate (orientation 6 rotates 90 degrees counterclockwise) with extra rotation when the document appears upside down. OCR tries macOS Vision first, then EasyOCR, then Tesseract if available, extracting text and identifying document types such as deposit, employment, or retirement certificates. Translation keeps structure, uses visa-appropriate terminology, preserves names with English in parentheses, formats Chinese names in pinyin, and keeps numbers, dates, and amounts accurate. A Python script using PIL and reportlab builds page one with the rotated original centered on A4 and page two with a bold centered title and left-aligned English body plus a certified translation note. Output is saved as original_filename_Translated.pdf beside the source. Supported inputs include bank deposit, income, employment, retirement, and property certificates, business licenses, IDs, and passports for Australia, USA, Canada, a.5.6kinstalls 307Plankton Code QualityThe plankton-code-quality skill documents Plankton, a write-time code quality enforcement system for Claude Code credited to @alxfazio. On every file edit, the multi_linter.sh PostToolUse hook runs three phases: silent auto-format with ruff format, biome, shfmt, taplo, and markdownlint fixing roughly forty to fifty percent of issues; JSON violation collection with line, column, code, message, and linter fields; then a claude -p subprocess that routes fixes to Haiku, Sonnet, or Opus by violation complexity with verify reruns. The main agent sees nothing when clean or fixed, or hook violation counts when fixes remain. Config protection uses PreToolUse blocks on linter configs, a stop hook git diff guardian, and protects .ruff.toml, biome.json, and related files. Bash PreToolUse blocks legacy pip and npm package managers favoring uv and bun. Setup requires manual install with brew install jaq ruff uv, uv sync --all-extras, and copying .claude hooks plus settings.json for per-project use across Python, TypeScript, Shell, YAML, JSON, TOML, Markdown, and Dockerfile.5.5kinstalls 308BlueprintThe blueprint skill from Everything Claude Code turns a one-line goal into a step-by-step build plan any coding agent can execute cold-start without prior session context. Use it to split large features into multiple PRs with clear dependency order, plan cross-session refactors or migrations, coordinate parallel sub-agent work, or avoid rework from context loss between sessions. Skip it for single-PR tasks, work finishable in under three tool calls, or when the user says just do it. The five-stage pipeline runs Research with git and gh preflight, Design splitting goals into 3-12 PR-sized steps with dependency edges, Draft writing self-contained markdown plans in plans/ with verification commands, Review via adversarial sub-agent against anti-pattern catalog, and Register persisting the plan. When git and GitHub CLI are available, Blueprint generates full branch, PR, and CI workflow plans; otherwise it switches to direct mode. Example: /blueprint myapp migrate database to PostgreSQL writes plans/myapp-migrate-database-to-postgresql.md with five migration steps. Ships with ECC; no separate install if ECC is present.5.5kinstalls 309Ggetgget is an ECC community skill for developers and bioinformatics engineers who need fast, reproducible lookups against public genome databases. It walks agents through Ensembl ID searches, transcript metadata, sequence retrieval, BLAST- or BLAT-style queries, reference genome links, and pathway or disease module checks using the gget CLI or Python interface inside a clean virtual environment. The workflow emphasizes logging first-pass evidence before graduating to heavier stacks such as Biopython, Snakemake, Nextflow, or BLAST+ with local indexes. Use gget for exploratory genomics questions, annotation pulls, and quick validation—not regulated clinical interpretation or high-throughput production pipelines requiring pinned database versions.5.5kinstalls 310Ai Regression TestingThe ai-regression-testing skill defines regression test patterns for AI-assisted development where the same model writes and reviews code, creating systematic blind spots only automated tests can catch. Activate when AI agents change API routes or backend logic, after bugs are fixed and need recurrence prevention, when sandbox or mock modes enable DB-free API tests, before running /bug-check reviews, or when sandbox versus production code paths diverge. Core failure loop: AI writes a fix, AI reviews it, declares correct, bug remains. Production examples show notification_settings added without SELECT updates, sandbox paths omitted after production fixes, and tests finally catching the fourth recurrence. Vitest plus Next.js App Router setup forces SANDBOX_MODE with createTestRequest helpers. Write tests for discovered bugs not hypothetical coverage, assert API response shape, parity test sandbox versus production fields, and run npm run test and build as mandatory first bug-check steps before AI review.5.5kinstalls 311Uspto Databaseuspto-database is a community-origin agent skill for querying United States Patent and Trademark Office systems from coding agents. The workflow prioritizes official USPTO surfaces—Open Data Portal datasets, Patent File Wrapper bibliographic records, PatentSearch API for granted patents and pre-grant publications, and TSDR for trademark status—then documents findings in reproducible investigation logs. Reach for uspto-database when searching granted patents, checking application status, reviewing assignment chains, or comparing USPTO records against Google Patents, Lens.org, or Semantic Scholar. The skill is for data collection and record verification only, not legal advice.5.5kinstalls 312Exa SearchExa Search is an agent skill that connects Claude Code and similar agents to the Exa MCP server for neural web search, filtered queries, code context, company intelligence, people lookup, URL crawling, and asynchronous deep research. Configure exa-mcp-server in your Claude settings with an EXA_API_KEY from exa.ai, then call web_search_exa for broad current-information queries, web_search_advanced_exa when you need includeDomains, excludeDomains, or published-date filters, and get_code_context_exa to pull snippets from GitHub, Stack Overflow, and documentation sites with adjustable tokensNum. company_research_exa and people_search_exa support business due diligence and professional profile discovery, while crawling_exa extracts full page text from URLs you already found. deep_researcher_start launches a background synthesis job you poll with deep_researcher_check while continuing other work. The skill documents defaults such as numResults and token budgets so agents choose focused snippets or wider context deliberately. Activate it when a developer says search, look up, find, or asks for the latest on a topic during implementation, competitive analysis, or technical investigation.5.5kinstalls 313Kotlin TestingThe kotlin-testing skill teaches comprehensive Kotlin testing with Kotest, MockK, coroutine testing, property-based testing, and Kover coverage following TDD methodology. Use when writing new Kotlin functions, adding coverage, implementing property tests, following red-green-refactor workflows, or configuring Kover thresholds. Workflow: identify target code, choose a Kotest spec style such as StringSpec, FunSpec, BehaviorSpec, or DescribeSpec, mock dependencies with MockK, run failing tests, implement minimal code, refactor, then verify coverage. Covers matchers, coEvery and coVerify for suspend functions, runTest with advanceTimeBy, Flow debounce testing, property forAll generators, withData tables, beforeSpec fixtures, spyk partial mocks, and Ktor testApplication HTTP tests. Kover enforces 80 percent minimum via koverVerify in CI with HTML and XML reports. Best practices mandate tests first, one Kotest style per project, behavior over implementation, and no Thread.sleep in coroutine tests.5.5kinstalls 314Benchmark Optimization Loopbenchmark-optimization-loop is an ECC-origin Claude Code skill that replaces open-ended requests like make it 20x faster with structured, measured optimization cycles. Before any change, the skill requires a defined operation, a correctness gate that must stay green, and a chosen metric such as wall time, p95 latency, rows per second, cost per run, or memory usage. Developers reach for benchmark-optimization-loop when comparing implementation variants, running recursive optimization attempts, or choosing the best approach through repeated measured tests rather than intuition. The skill uses Read, Write, Edit, Bash, Grep, and Glob tools to iterate on code while holding benchmarks accountable. It fits backend services, data pipelines, and CLI tools where performance regressions are costly.5.5kinstalls 315Pytorch PatternsPyTorch Development Patterns teaches idiomatic PyTorch for robust, efficient, and reproducible deep learning applications. Activate when writing models or training scripts, reviewing deep learning code, debugging training loops or data pipelines, optimizing GPU memory or training speed, or setting up reproducible experiments. Core principles demand device-agnostic code with torch.device and .to(device), full seed control across torch, numpy, and cudnn, and explicit tensor shape documentation in forward passes. Model patterns favor clean nn.Module structure, explicit Kaiming initialization, and separating features from classifiers. Training loops set model.train() and model.eval(), use optimizer.zero_grad(set_to_none=True), mixed precision with torch.amp.autocast and GradScaler, gradient clipping, and @torch.no_grad() evaluation. Data pipelines use typed Dataset classes, optimized DataLoader settings with pin_memory and persistent workers, and collate_fn for variable-length batches. Checkpointing saves model, optimizer, epoch, and loss state. Performance options include gradient checkpointing, torch.compile, and profiling with torch.profiler.5.5kinstalls 316Latency Critical Systemslatency-critical-systems is an ECC-origin skill from affaan-m/ecc for engineering latency-sensitive systems such as realtime dashboards, market data feeds, streaming agents, execution gateways, queues, caches, and HFT-like infrastructure. The skill instructs agents to split metrics instead of collapsing everything into fast, tracking p50, p95, and p99 latency alongside throughput and data freshness on hot paths. It explicitly focuses on engineering patterns and does not authorize live trading or financial advice. Developers reach for latency-critical-systems when users care about realtime behavior, streaming freshness, or execution speed in production services. Tooling access includes Read, Write, Edit, Bash, Grep, and Glob for profiling codepaths, queue backlogs, and cache invalidation strategies across backend services.5.4kinstalls 317Documentation LookupDocumentation Lookup is an agent skill that routes library and framework questions to live documentation through the Context7 MCP instead of relying on training data. Activate when users ask setup or configuration questions, request code that depends on a specific library, need API or reference information, or name frameworks such as React, Next.js, Prisma, Supabase, Vue, Svelte, Express, or Tailwind. The workflow resolves a Context7-compatible library ID with resolve-library-id using the library name and full user question, selects the best match by name fit benchmark score source reputation and version specificity, then calls query-docs with that library ID and the precise task query. Answers should cite the library or version when it matters and include relevant code snippets from fetched docs. Hard limit: no more than three resolve-library-id or query-docs calls per question; if still unclear, state uncertainty rather than guessing. Redact API keys passwords tokens and secrets before sending queries to Context7.5.4kinstalls 318Flutter Dart Code Reviewflutter-dart-code-review is a library-agnostic Flutter and Dart code review checklist from Everything Claude Code. It spans project health, Dart language pitfalls, widget decomposition, const and key usage, theme tokens, state management for BLoC, Riverpod, Provider, GetX, MobX, and Signals, performance rebuild boundaries, testing targets, accessibility semantics, platform concerns, secure storage, package vetting, navigation typing, error reporting, l10n, DI, and static analysis. Reviewers confirm widgets stay presentation-only, business logic lives outside build methods, sealed or union async states replace boolean isLoading flags, subscriptions dispose cleanly, ListView.builder handles large lists, HTTPS and Keychain storage protect secrets, routes stay typed, and business logic targets eighty percent line coverage. Principles apply regardless of routing library, DI framework, or state solution. Use before merging Flutter PRs, auditing mobile quality, or aligning reviewers on Dart idioms and widget performance.5.4kinstalls 319Customs Trade ComplianceThis customs and trade compliance skill encodes fifteen plus years of senior trade compliance expertise across US, EU, UK, and Asia-Pacific jurisdictions. Activate when classifying products under HS or HTS tariff codes, preparing customs entry documents such as commercial invoices and certificates of origin, screening parties against restricted entity lists like OFAC SDN and EU sanctions, evaluating FTA eligibility and duty reduction opportunities, or responding to customs audits and penalty notices. The documented six-step workflow classifies products using GRI rules and chapter analysis, determines applicable duty rates and preferential programs including FTZ and FTA, screens all transaction parties before shipment, prepares and validates entry documents per jurisdiction, monitors tariff and sanction regulatory changes, and responds to government inquiries with voluntary disclosure and penalty mitigation strategies. Examples cover HS reclassification disputes, USMCA regional value content analysis, and restricted party screening false-positive resolution. Integrates with ACE, CHIEF/CDS, ATLAS, broker portals, and ERP trade management modules.5.4kinstalls 320Parallel Execution Optimizerparallel-execution-optimizer is an ECC agent skill for safely speeding up large coding tasks by planning parallel execution lanes before any tool calls run. The workflow starts with goals and completion signals, splits work into channels labeled parallel, serial, or gated, batches independent reads and checks, and isolates writes by file, git worktree, branch, service, or dataset. Developers reach for parallel-execution-optimizer when repo scans, API checks, browser checks, build or test runs, and multi-worktree implementation batches can run at the same time without stepping on each other. The skill closes with a verification table that proves each lane succeeded instead of relying on a vague speed claim. It uses Read, Write, Edit, Bash, Grep, and Glob tools from the ECC toolkit.5.4kinstalls 321Social Publishersocial-publisher is a community-origin ECC skill that connects Claude Code and compatible agents to SocialClaw at getsocialclaw.com for agent-driven social media publishing. A single workspace API key enables scheduling and posting across 13 platforms: X, LinkedIn, Instagram, Facebook Pages, TikTok, Discord, Telegram, YouTube, Reddit, WordPress, and Pinterest. Developers use social-publisher to publish or schedule content, upload media, manage campaigns, and monitor post delivery status without leaving the agent chat. The skill activates when users want multi-platform distribution or campaign orchestration rather than drafting copy alone. It is an integration workflow over SocialClaw's workspace API rather than a standalone social network client.5.4kinstalls 322Data Scraper Agentdata-scraper-agent builds production-ready automated AI data collection agents for any public source including job boards, prices, news, GitHub, and sports scores. The three-layer architecture follows COLLECT, ENRICH, STORE: Playwright or BeautifulSoup scrapers run on schedule, Gemini Flash scores and classifies extracted text, and results land in Notion, Google Sheets, or Supabase. GitHub Actions provides free cron scheduling with no paid hosting required. Workflow steps define the source and fields, implement the scraper collector, configure Gemini enrichment, set storage, wire the schedule, and add a feedback loop from user judgments. Activate when users want to scrape or monitor public websites, track jobs, prices, news, repos, sports scores, events, or lists, or ask how to automate collection without paying for hosting. Examples include job board monitoring for new postings. The stack is Python with open-source scraping, free LLM enrichment, and free tier storage backends.5.4kinstalls 323Enterprise Agent OpsEnterprise Agent Ops guides teams running cloud-hosted or continuously running agent systems that need operational controls beyond single CLI sessions. The skill organizes work across four domains: runtime lifecycle with start, pause, stop, and restart; observability through logs, metrics, and traces; safety controls including scopes, permissions, and kill switches; and change management covering rollout, rollback, and audit. Baseline controls require immutable deployment artifacts, least-privilege credentials, environment-level secret injection, hard timeout and retry budgets, and audit logs for high-risk actions. Track success rate, mean retries per task, time to recovery, cost per successful task, and failure class distribution. When failure spikes, freeze new rollout, capture representative traces, isolate the failing route, patch with the smallest safe change, run regression and security checks, then resume gradually. Deployment integrations pair with PM2 workflows, systemd services, container orchestrators, and CI/CD gates.5.4kinstalls 324Ralphinho Rfc PipelineRalphinho RFC Pipeline is an agent skill that decomposes oversized features into independently verifiable work units using Humanplane-style RFC breakdown and multi-unit orchestration. Activate when a single-agent path cannot safely deliver a large capability and you need DAG-based decomposition, quality gates, and a disciplined merge queue. The seven-stage pipeline runs RFC intake, DAG decomposition, unit assignment, implementation, validation, merge queue integration, and final system verification. Each unit ships with id, depends_on, scope, acceptance_tests, risk_level, and rollback_plan. Complexity tiers classify isolated file edits as Tier 1, multi-file behavior as Tier 2, and schema, auth, performance, or security work as Tier 3. Every unit passes research, implementation plan, implementation, tests, review, and merge-ready reporting. Merge rules forbid integrating units with unresolved dependency failures, require rebasing on the latest integration branch, and mandate integration test reruns after each queued merge. Stalled units are evicted, findings snapshotted, scope narrowed, and retries constrained. Outputs include RFC execution logs, unit scorecards, dependency graph s.5.4kinstalls 325Data Throughput Acceleratordata-throughput-accelerator is an ECC-origin skill for when moving, transforming, or saving large datasets is the bottleneck in production pipelines. It forces an upfront split between source extraction, network transfer, warehouse load, transform, serving-table freshness, and proof of correct landing before applying parallelization or batching tactics.py The workflow targets backfills, exports, manifest catch-up, and table sync jobs where faster wrong data is unacceptable. Data engineers and backend developers reach for data-throughput-accelerator when ingestion lag, warehouse load queues, or sync freshness SLAs break and they need a structured acceleration plan with evidence that rows landed correctly.5.4kinstalls 326Prediction Market Oracle ResearchPrediction-market-oracle-research is an affaan-m ECC origin skill for source-grounded analysis of prediction markets as forecasting inputs, oracle-like signals, or decision-intelligence layers. It guides developers through evaluating market-implied probabilities while separating venue mechanics, liquidity constraints, and integration caveats from treating prices as objective truth. Explicit guardrails prohibit investment advice and trading recommendations. Reach for this skill when building agents, dashboards, or corporate decision systems that might ingest prediction-market data and need rigorous due diligence on signal quality, not speculative trading guidance. The research output covers integration patterns and documented limitations for engineering teams.5.4kinstalls 327Prediction Market Risk Reviewprediction-market-risk-review is an ECC-origin agent skill that gates prediction-market and trading automation workflows before they handle venue authentication, user portfolio data, API keys, or trade planning. Review gates cover advice boundaries—keeping outputs informational without buy/sell/hold sizing recommendations—and venue plus regulatory concerns alongside data-quality and privacy checks. Developers invoke prediction-market-risk-review when building agents that interact with prediction markets, oracle feeds, basket strategies, or execution-capable trading tools and need a structured preflight audit. The skill emphasizes explicit manual decision points and removal of direct trading recommendations from agent outputs before any workflow touches user financial context.5.4kinstalls 328Recursive Decision Ledgerrecursive-decision-ledger is a Claude Code skill from affaan-m/ecc for agents that must explore multiple options, run repeated trials, or perform recursive reasoning with a visible evidence trail. It preserves repeated rollouts, prior memory, fresh information, and explicit marks while avoiding false certainty from looping alone. The ledger contract records each roll's decisions and outcomes for stochastic optimization, local-optima exploration, ensemble comparison, and high-dimensional search tasks. Developers reach for recursive-decision-ledger when prompting agents through Prime Gauss-style recursive processes that need auditable marks instead of opaque iteration. Tools include Read, Write, Edit, Bash, Grep, and Glob.5.3kinstalls 329Kotlin Ktor Patternskotlin-ktor-patterns catalogs production layouts and plugin recipes for Kotlin coroutine Ktor HTTP servers. A standard project splits Application.kt module wiring from plugins for Routing, Serialization, Authentication, StatusPages, and CORS, with routes, models, services, repositories, and Koin DI under com.example. Routing DSL examples implement REST userRoutes using call.receive and call.respond, JWT-protected authenticate blocks, and authRoutes for login, register, and me endpoints. Serialization installs ContentNegotiation with kotlinx.serialization Json settings, ApiResponse envelopes, PaginatedResponse, and a custom InstantSerializer. Authentication configures JWT verifier, audience checks, realm, and Unauthorized challenges. StatusPages maps ContentTransformationException, AuthenticationException, and NotFoundException to typed ApiResponse errors. Koin modules wire repositories and services with inject in routes plus FunSpec KoinTest units. WebSocket chatRoutes snapshot connections under lock to avoid ConcurrentModificationException. testApplication tests cover unauthenticated POST rejection and bearerAuth success. Invoke when building Ktor REST APIs, configuring plugins,.5.3kinstalls 330Context BudgetThe context-budget skill audits Claude Code context window consumption across agents, skills, MCP servers, rules, and CLAUDE.md files, then surfaces prioritized token savings. Run it when session performance or output quality drops, after adding many skills agents or MCP servers, before planning new components, or via the /context-budget command. Phase one inventories token overhead with word-count times 1.3 heuristics, flags heavy agents over 200 lines, bloated descriptions over 30 words, skills over 400 lines, rules over 100 lines, and MCP servers with 20+ tools or CLI-replaceable wrappers. Phase two buckets components as always needed, sometimes needed, or rarely needed. Phase three detects bloated agent descriptions, redundant skills duplicating agents, MCP oversubscription past ten servers, and CLAUDE.md bloat. Phase four emits a Context Budget Report with overhead totals, component breakdown, ranked issues, and top three optimizations. Verbose mode adds per-file token counts and MCP tool schema size estimates. Docs note MCP tool schemas cost roughly 500 tokens each.5.3kinstalls 331Nanoclaw ReplNanoClaw REPL is an ECC skill for running and extending scripts/claw.js, the zero-dependency session-aware REPL in NanoClaw v2 built on claude -p. Invoke it when you operate the REPL or add command handlers while keeping markdown-as-database compatibility. Persistent markdown-backed sessions survive across runs. Slash commands include model switching with /model, dynamic skill loading with /load, session branching with /branch, cross-session search with /search, history compaction with /compact, export to md, json, or txt with /export, and session metrics with /metrics. Operating guidance keeps sessions task-focused, branches before high-risk changes, compacts after major milestones, and exports before sharing or archival. Extension rules require zero external runtime dependencies, preserved markdown storage, and deterministic local command handlers when modifying claw.js. The skill documentation targets developers maintaining NanoClaw workflows inside Everything Claude Code.5.3kinstalls 332Inventory Demand PlanningInventory demand planning codifies senior retail demand planner expertise for multi-location retailers with 40-200 stores and 300-800 active SKUs across grocery, general merchandise, seasonal, and promotional assortments. Use it when generating or reviewing demand forecasts, setting safety stock from demand variability and service targets, planning replenishment for seasonal transitions or promotions, evaluating forecast accuracy with MAPE, WMAPE, and bias, and making buy decisions under supplier MOQ or lead time changes. The workflow collects POS sell-through and shipment signals, selects forecasting methods per ABC/XYZ classification, calculates safety stock with service-level z-scores, and generates suggested purchase orders with MOQ rounding. Coverage includes promotional lift estimation, cannibalization offsets, post-promo dip forecasting, seasonal open-to-buy timing, markdown triggers, and escalation protocols for stockouts or vendor lead time shocks. Pair with planning suites such as Blue Yonder, Oracle Demantra, or Kinaxis plus ERP and WMS context when translating merchandising intent into executable POs while minimizing stockouts and excess inventory.5.3kinstalls 333Quality NonconformanceThe quality-nonconformance skill encodes senior quality engineering expertise for regulated manufacturing under FDA 21 CFR 820, IATF 16949, AS9100, and ISO 13485. Activate it when investigating NCRs from incoming, in-process, or final inspection, performing root cause analysis with 5-Why, Ishikawa, or fault tree methods, deciding material disposition, drafting or reviewing CAPA plans, interpreting SPC control chart signals, or responding to regulatory audit findings. The six-step workflow covers detection, containment, severity classification, structured root cause investigation, disposition under engineering and regulatory constraints, and corrective action verification with documented closure. Examples walk through incoming AQL failures with supplier SCARs, Western Electric Rule 2 SPC signals where in-spec differs from in-control, and automotive OEM 8D CAPA for field failures. Core knowledge spans the full NCR lifecycle with identification, quarantine, MRB segregation, and traceability tied to eQMS platforms, ERP QM modules, Minitab, and supplier portals.5.3kinstalls 334Mcp Server PatternsThe mcp-server-patterns skill guides developers building Model Context Protocol servers so AI assistants can call tools, read resources, and surface prompts from custom backends. It covers core MCP concepts including tool registration for invocable actions, read-only resources with URI handlers, reusable prompt templates, and transport choice between stdio for local clients like Claude Desktop and Streamable HTTP for remote clients such as Cursor or cloud hosts. The Node and TypeScript SDK API evolves across versions, so agents must verify current registerTool, tool, registerResource, and connect signatures against official MCP docs or Context7 instead of copying stale examples. Best practices emphasize schema-first tool inputs with Zod validation, structured errors the model can interpret, idempotent tools for safe retries, and pinned SDK versions with release-note checks on upgrade. Examples show installing @modelcontextprotocol/sdk with zod, instantiating McpServer, and keeping server logic transport-independent so stdio or HTTP entrypoints can be swapped. Use when implementing a new MCP server, adding tools or resources, choosing transport, upgrading the SDK, or debugging regi.5.3kinstalls 335Production SchedulingThe production-scheduling skill encodes senior scheduler expertise for discrete and batch manufacturing lines with ERP, finite-capacity APS, MES, and CMMS coordination. It applies when production orders compete for constrained work centers, disruptions require rapid re-sequencing, changeover campaign trade-offs need economic decisions, or hot orders must slot into frozen schedules. The workflow identifies the system constraint with OEE and utilization data, classifies demand by past-due and constraint-feeding priority, sequences jobs with EDD, SPT, or setup-aware rules, optimizes changeovers via setup matrices with nearest-neighbor and 2-opt improvement, and locks a 24 to 48 hour stabilization window to limit churn. Disruption response re-plans only unlocked jobs and publishes updates to MES. Frameworks include TOC drum-buffer-rope, SMED, OEE analysis, forward versus backward scheduling, and ERP/MES interaction patterns informed by 15+ years of scheduler practice. Examples cover CNC breakdown rerouting, campaign versus mixed-model crossover analysis, and rush order insertion without breaking frozen commitments.5.3kinstalls 336Energy ProcurementThis energy procurement skill encodes senior commercial and industrial energy manager expertise for organizations spending fifteen to eighty million dollars annually across ten to fifty plus facilities in regulated and deregulated markets. Activate when running electricity or natural gas RFPs, analyzing tariff structures and rate schedule optimization, managing demand charge mitigation, evaluating physical or virtual PPAs, building annual energy budgets with layered hedging, or responding to market volatility such as polar vortex events or regulatory rate cases. The documented workflow profiles facility load shapes from interval meter data, structures supplier RFPs across fixed index and block-and-index products, evaluates total cost of energy including capacity and transmission, executes staggered contracts to avoid concentration risk, and monitors hedge positions with monthly variance reporting. Coverage spans utility bill anatomy including energy demand and capacity charges, layered procurement and demand response programs, renewable REC and PPA economics, load profiling and market structure differences across PJM ERCOT and CAISO, Scope 2 sustainability reporting, and escalatio.5.3kinstalls 337Carrier Relationship ManagementCarrier-relationship-management encodes a senior transportation manager playbook for portfolios of forty to two hundred active carriers across truckload, LTL, intermodal, and brokerage modes. It covers sourcing and vetting carriers through FMCSA SAFER, insurance verification, annual and lane-specific RFPs with volume commitments, and negotiation of linehaul, fuel surcharge tables, accessorial schedules, and minimum charges. The skill explains contract versus spot mix targets of seventy-five to eighty-five percent contract freight, routing guide construction with primary and backup assignments, and weighted scorecards tracking on-time performance, claims rate, tender acceptance, and cost. Workflows include onboarding regional LTL carriers with ninety-day trial scorecards, running two-hundred-lane TL RFPs against DAT benchmarks, and reassigning freight when tender acceptance drops during tight capacity. Systems referenced include TMS, rate management platforms, DAT or Greenscreens market data, and FMCSA compliance checks.5.3kinstalls 338Fal Ai MediaThe fal-ai-media skill routes unified media generation through the fal.ai MCP server with search, find, generate, result, status, cancel, estimate_cost, models, and upload tools. Configure fal-ai in Claude settings with npx fal-ai-mcp-server and a FAL_KEY from fal.ai. Image workflows cover Nano Banana 2 for fast drafts and Nano Banana Pro for production fidelity with image_size, num_images, seed, and guidance_scale parameters, plus image editing via upload and image_url inputs. Video generation documents Seedance 1.0 Pro, Kling Video v3 Pro, and Veo 3 with duration, aspect_ratio, and image-to-video patterns. Audio covers CSM-1B text-to-speech and ThinkSound video-to-audio, with optional ElevenLabs REST or VideoDB generative audio when configured. Tips recommend starting with lower-cost Nano Banana 2 for prompt iteration, using seed for reproducibility, and calling estimate_cost before expensive video runs in production.5.3kinstalls 339CrosspostThe crosspost skill distributes one underlying idea across X, LinkedIn, Threads, and Bluesky while preserving the author's voice and platform constraints. Core rules forbid publishing identical copy, inventing CTAs or closing questions the source did not earn, and padding LinkedIn posts with fake founder-reflection tone. Workflow starts from the strongest source version, optionally runs brand-voice for a reusable VOICE PROFILE and content-engine for voice shaping, then adapts per platform: compressed claims on X, necessary context on LinkedIn without forced professionalism, direct readable Threads copy, and concise Bluesky posts without hashtag gaming. Default posting order publishes the strongest native version first, then secondary platforms, staggering timing only when requested. Banned patterns include Excited to share, Here's what I learned, What do you think, and link in bio unless literally true. Output returns primary version, adapted variants, change notes, and unresolved publishing constraints.5.2kinstalls 340Logistics Exception ManagementGet expert, codified guidance when freight shipments go wrong.5.2kinstalls 341VideodbThe videodb skill provides perception, memory, and actions for video, live streams, and desktop sessions through the VideoDB Python SDK. Setup installs videodb with capture extra via pip, loads VIDEO_DB_API_KEY from environment or project .env, and connects with videodb.connect(). Workflows cover uploading files, URLs, YouTube links, or RTSP feeds; building spoken, visual, and keyword indexes; searching moments with timestamps and auto-clips; transcoding codec, fps, resolution, and aspect ratio; and timeline edits for subtitles, overlays, dubbing, and generative audio. Desktop capture starts screen, mic, and system audio sessions with real-time alerts and searchable episodic memory. Search examples wrap InvalidRequestError for empty results, use index_scenes with score_threshold 0.3 for semantic scene search, and compile evidence stream URLs. Agents should use inline python or heredocs for short commands, never handle API keys directly, and obtain free keys at console.videodb.io with fifty free uploads.5.2kinstalls 342Agent EvalThe agent-eval skill provides a lightweight CLI to compare coding agents like Claude Code, Aider, and Codex on reproducible custom tasks instead of subjective preference. Tasks are declarative YAML files specifying actions, files touched, and deterministic judges for success. Each agent run gets an isolated git worktree without Docker so agents cannot interfere with each other or corrupt the base repository. Collected metrics include pass rate against judges, API cost per task when available, wall-clock completion time, and consistency across repeated trials such as three-of-three passes. Workflow creates a tasks directory, runs agents against tasks, generates comparison reports, and supports judge types: codebase deterministic tests, pattern-based checks, and optional LLM-as-judge with caution about noise. Best practices recommend starting with three to five real workload tasks, at least three trials per agent, pinning task YAML commits for reproducibility, including at least one deterministic judge per task, tracking cost alongside pass rate, and versioning task definitions like test fixtures.5.2kinstalls 343Kotlin Exposed PatternsThe kotlin-exposed-patterns skill documents JetBrains Exposed database access for Kotlin services using DSL and DAO styles, HikariCP connection pooling, Flyway versioned migrations, and repository boundaries testable with in-memory H2. Exposed offers SQL-like DSL expressions and entity lifecycle DAO APIs; HikariConfig configures reusable pools; Flyway runs versioned SQL at startup; operations run inside newSuspendedTransaction for coroutine-safe atomicity. Examples span Table definitions with UUIDTable, selectAll with where filters, insertAndGetId, update, deleteWhere, batchInsert, joins, orderBy limit offset, and aggregate count sum avg functions. DAO sections cover entity definitions, DAO operations, suspend transaction support, and isolation levels. Repository pattern separates business logic from data access. JSON columns use kotlinx.serialization JSONB patterns. Quick reference maps object Table UUIDTable, newSuspendedTransaction blocks, innerJoin leftJoin, and batchInsert for bulk efficiency in production Kotlin API and coroutine-backed services.5.2kinstalls 344Motion Uimotion-ui is a skill from affaan-m/ecc in the Everything Claude Code collection that helps teams define a coherent Motion-based animation system for React and Next.js applications. The public readme indicates a Japanese translation is in progress and references the original English SKILL.md for the full motion-ui guidance on accessibility, performance, and reusable transition patterns. Developers reach for motion-ui when adding animations without introducing janky interactions, inconsistent timing, or accessibility regressions across components. motion-ui pairs naturally with motion-advanced when a project needs a solid baseline motion language before layering complex gestures or SVG path effects.5.2kinstalls 345Perl PatternsThe perl-patterns skill teaches idiomatic Modern Perl 5.36+ for robust maintainable applications with a bias toward signatures, explicit modules, centralized error handling, and testable boundaries. Core principles start with use v5.36 replacing strict warnings and subroutine signatures, understanding scalar versus list context, postfix dereference readability, and the isa infix type check from 5.32. Error handling covers eval die, Try::Tiny, and native try catch in 5.40. OO prefers lightweight Moo with roles, reserving Moose only when metaprotocols are required, plus native class keyword notes for 5.38 Corinna. Regex guidance promotes named captures, possessive quantifiers, and timeout guards against ReDoS. File I/O uses three-argument open and Path::Tiny. Module layout follows standard project structure with exporter patterns, perltidy and perlcritic rc files, and cpanfile plus carton dependency management. Activate when writing new Perl modules, reviewing idioms, refactoring legacy Perl, designing module architecture, or migrating pre-5.36 code.5.2kinstalls 346Returns Reverse LogisticsReturns-reverse-logistics encodes a senior returns operations manager playbook for retail and ecommerce reverse logistics across RMA authorization, receiving and inspection, grade assignment, disposition routing, refunds, fraud detection, vendor chargebacks, and warranty claims. Systems include OMS, WMS, RMS, CRM, fraud platforms, and vendor portals. Workflows receive return requests, verify policy windows and category restrictions, issue prepaid labels or drop-off instructions, inspect received units with A-through-D grades, route to resale refurb clearance destroy or vendor return channels based on recovery economics, process refunds or exchanges, flag anomalies for fraud review, and aggregate vendor-chargeable RTV claims within contractual windows. Core knowledge covers standard thirty-day return windows, shorter fifteen-day electronics rules, restocking fees on opened electronics, cross-channel BORIS online-buy store-return pricing against original purchase price, international duty drawback constraints, and exception frameworks for price-match returns or loyalty-tier extensions. Examples walk high-value laptop fraud inspection, serial returner pattern analysis at forty-seven.5.2kinstalls 347Dmux WorkflowsThe dmux-workflows skill orchestrates parallel AI agent sessions using dmux, a tmux pane manager where n creates a new pane with a prompt and m merges pane output back to the main session. Install via npm install -g dmux from github.com/standardagents/dmux. Supported harnesses include Claude Code, Codex, OpenCode, Cline, Gemini, and Qwen. Patterns cover research plus implement split panes, multi-file feature parallelism across schema API and UI tracks, test watcher plus fixer loops, cross-harness reviews mixing Claude Code and Codex, and parallel security performance coverage review pipelines. Best practices require independent tasks without cross-pane dependencies, clear file or concern boundaries, strategic merge review before integration, git worktrees when file conflicts are likely, and keeping total panes under five to six for token cost control. Git worktree integration shows feat branch worktrees for auth and billing parallel development. Activate when users say run in parallel, split this work, use dmux, or multi-agent.5.2kinstalls 348Claude DevfleetThe claude-devfleet skill provides agent guidance for claude devfleet tasks documented in ECC everything-claude-code. It defines activation triggers, role context, step-by-step workflows, examples, and domain checklists agents should follow instead of improvising APIs or business rules. Use it when users ask for claude devfleet implementation, review, negotiation, testing, or orchestration patterns that match the skill name and upstream documentation. The skill encodes operational expertise, required tooling, decision frameworks, and quality gates so agents produce consistent outputs aligned with the source playbook. Refer to SKILL.md for prerequisites, command examples, and edge cases before running production workflows. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use.5.2kinstalls 349X ApiThe x-api skill documents programmatic X integration for posting, reading, searching, and analytics. OAuth 2.0 bearer tokens suit read-heavy search and public data via X_BEARER_TOKEN headers against api.x.com v2 endpoints. OAuth 1.0a user context with X_CONSUMER_KEY, X_CONSUMER_SECRET, X_ACCESS_TOKEN, and X_ACCESS_TOKEN_SECRET is required for posting tweets, threads, account management, and DMs using requests_oauthlib OAuth1Session. Core operations include single tweet POST, thread helper chaining reply in_reply_to_tweet_id, user timeline fetch with tweet.fields metrics, recent search with query operators, voice sampling search excluding retweets and replies, user lookup by username, and v1.1 media upload followed by v2 tweet with media_ids. Rate limits vary by endpoint and tier; read x-rate-limit-remaining and x-rate-limit-reset headers at runtime with backoff instead of hardcoded tables. Security rules forbid hardcoding tokens, committing .env files, and require rotation on exposure. Content integration flows through brand-voice and content-engine with draft approval before posting unless explicitly requested.5.2kinstalls 350Perl SecurityThe perl-security skill provides comprehensive Perl security guidance for web apps in CGI, Mojolicious, Dancer2, and Catalyst covering taint mode, input validation, safe execution, and injection prevention. Taint mode with perl -T tracks external data; sanitize PATH early and delete risky env vars like IFS and BASH_ENV. Untaint with specific capturing regexes, rejecting overly permissive .* patterns that defeat taint checks. Input validation prefers allowlists for sort fields and emails, length constraints on comments up to ten thousand characters, and ReDoS-safe regex with possessive quantifiers or match timeouts via alarm. Safe process execution avoids shell interpolation; DBI queries must use parameterized placeholders never string concatenation. Web sections address XSS escaping, CSRF tokens, Content-Security-Policy headers, and session fixation risks. perlcritic security themes automate policy checks. Activate when handling user input, reviewing Perl vulnerabilities, performing path operations on user-supplied paths, executing system commands from Perl, or writing DBI queries touching untrusted data.5.2kinstalls 351Perl TestingThe perl-testing skill provides agent guidance for perl testing tasks documented in ECC everything-claude-code. It defines activation triggers, role context, step-by-step workflows, examples, and domain checklists agents should follow instead of improvising APIs or business rules. Use it when users ask for perl testing implementation, review, negotiation, testing, or orchestration patterns that match the skill name and upstream documentation. The skill encodes operational expertise, required tooling, decision frameworks, and quality gates so agents produce consistent outputs aligned with the source playbook. Refer to SKILL.md for prerequisites, command examples, and edge cases before running production workflows. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use.5.2kinstalls 352Team BuilderThe team- skill provides agent guidance for team tasks documented in ECC everything-claude-code. It defines activation triggers, role context, step-by-step workflows, examples, and domain checklists agents should follow instead of improvising APIs or business rules. Use it when users ask for team implementation, review, negotiation, testing, or orchestration patterns that match the skill name and upstream documentation. The skill encodes operational expertise, required tooling, decision frameworks, and quality gates so agents produce consistent outputs aligned with the source playbook. Refer to SKILL.md for prerequisites, command examples, and edge cases before running production workflows. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use.5.2kinstalls 353Jira IntegrationThe jira-integration skill connects AI coding workflows to Jira for ticket retrieval, requirement analysis, progress comments, status transitions, and JQL search. Option A recommends the mcp-atlassian MCP server via uvx with JIRA_URL, JIRA_EMAIL, and JIRA_API_TOKEN in MCP env blocks never committed to source. Option B uses direct REST API v3 with a jira_curl helper that passes credentials via stdin to avoid command-line exposure. MCP tools include jira_search, jira_get_issue, jira_create_issue, jira_update_issue, jira_transition_issue, jira_add_comment, jira_get_sprint_issues, jira_create_issue_link, and jira_get_issue_development_info. Ticket analysis extracts testable requirements, acceptance criteria, test types, edge cases, and structured output templates. Update guidance maps workflow steps like branch creation, PR links, and merge to Jira comments and transitions. Security rules forbid hardcoded tokens, require env or secret manager storage, and document 401, 403, 404, timeout, and uvx PATH troubleshooting steps.5.1kinstalls 354Design SystemThe design-system skill provides agent guidance for design system tasks documented in ECC everything-claude-code. It defines activation triggers, role context, step-by-step workflows, examples, and domain checklists agents should follow instead of improvising APIs or business rules. Use it when users ask for design system implementation, review, negotiation, testing, or orchestration patterns that match the skill name and upstream documentation. The skill encodes operational expertise, required tooling, decision frameworks, and quality gates so agents produce consistent outputs aligned with the source playbook. Refer to SKILL.md for prerequisites, command examples, and edge cases before running production workflows. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use. Refer to SKILL.md for prerequisites, examples, and guardrails before production use.5.1kinstalls 355Nuxt4 PatternsThe nuxt4-patterns skill documents Nuxt 4 app patterns for SSR, hybrid rendering, route rules, and page-level data fetching. Hydration safety requires deterministic first renders without Date.now, Math.random, browser-only APIs, or storage reads in SSR templates. Browser logic belongs in onMounted, import.meta.client, ClientOnly, or .client.vue components. Data fetching prefers await useFetch for SSR-safe reads that hydrate from the Nuxt payload, useAsyncData with stable keys for composed sources, and lazy true or useLazyFetch for non-blocking secondary data with server false only when SEO irrelevant. Route rules in nuxt.config.ts configure prerender, swr, isr, ssr false, cache, and redirect per route group. Performance guidance uses Lazy component prefix with v-if gating, hydrate-on-visible, defineLazyHydrationComponent strategies, and NuxtLink for internal navigation prefetch. Review checklist verifies matching SSR and client markup, no top-level $fetch for page data, lazy loading UI for non-critical data, aligned route rules, and deferred hydration for heavy islands.5kinstalls 356Browser QaThe browser-qa skill automates visual and interaction testing on live pages after staging or preview deploys using browser automation MCP such as claude-in-chrome, Playwright, or Puppeteer. Phase one smoke tests navigate the target URL, filter console errors, verify no 4xx or 5xx network failures, capture desktop and mobile screenshots, and check Core Web Vitals targets LCP under 2.5s, CLS under 0.1, and INP under 200ms. Phase two interaction tests click navigation links, submit valid and invalid forms, exercise auth login logout flows, and run critical user journeys like checkout or search. Phase three visual regression screenshots key pages at 375px, 768px, and 1440px breakpoints, comparing baselines and flagging layout shifts over 5px, missing elements, or overflow. Phase four accessibility runs axe-core or equivalent, flags WCAG AA violations, verifies keyboard navigation, and checks landmarks. Output follows a structured QA Report markdown template with pass fail sections per phase. Pairs with benchmark and canary-watch skills for frontend pull request workflows.5kinstalls 357Hexagonal ArchitectureThe hexagonal-architecture skill teaches ports and adapters to keep business logic independent from frameworks, transport, and persistence. Core concepts include domain models without framework imports, use cases orchestrating workflows, inbound ports as application commands, outbound ports for repositories and gateways, adapters at HTTP or DB edges, and a composition root binding concrete implementations. Dependency direction flows inward: adapters to application to domain. Six steps model a use case boundary, define outbound ports first, implement pure orchestration, build inbound and outbound adapters, and wire in composition root with slice-by-slice testing using fake ports. Suggested module layout organizes features by domain, application ports and use-cases, adapters, and composition containers. TypeScript examples show OrderRepositoryPort, PaymentGatewayPort, and CreateOrderUseCase with immutable markAuthorized transforms. Multi-language mapping covers TypeScript, Java, Kotlin, and Go package splits. Migration playbook uses strangler approach, facade-first legacy wraps, and characterization tests without big-bang rewrites.5kinstalls 358Rules DistillThe rules-distill skill maintains agent rule files by scanning installed skills, cross-reading them against existing rules, and proposing append, revise, new section, or new file updates with user approval only. Phase 1 runs deterministic scripts to inventory skills and rules headings. Phase 2 batches skills thematically and uses LLM judgment to extract candidates that appear in two or more skills, are actionable behavior changes with clear violation risk, and are not already covered. Verdicts include Append, Revise, New Section, New File, Already Covered, or Too Specific. Phase 3 presents a summary table and per-candidate evidence, draft text, and violation risk for user approve, modify, or skip actions. Never modifies rules automatically. Results save to results.json with UTC timestamps and kebab-case candidate IDs. Design principles emphasize what not how, link back to source skills, deterministic collection plus LLM judgment, and a three-layer anti-abstraction filter.5kinstalls 359Remotion Video CreationThe remotion-video-creation skill Best practices for Remotion - Video creation in React. 29 domain-specific rules covering 3D, animations, audio, captions, charts, transitions, and more.. When to use Use this skills whenever you are dealing with Remotion code to obtain the domain-specific knowledge. How to use Read individual rule files for detailed explanations and code examples: - [rules/3d.md](rules/3d.md) - 3D content in Remotion using Three.js and React Three Fiber - [rules/animations.md](rules/animations.md) - Fundamental animation skills for Remotion - [rules/assets.md](rules/assets.md) - Importing images, videos, audio, and fonts into Remotion - [rules/audio.md](rules/audio.md) - Using audio and sound in Remotion - importing, trimming, volume, speed, pitch - [rules/calculate-metadata.md](rules/calculate-metadata.md) - Dynamically set composition duration, dimensions, and props - [rules/can-decode.md](rules/can-decode.md) Agents should read SKILL.md quick start steps, verify required binaries and environment variables, and follow reference files for exact parameters before calling tools.5kinstalls 360Git WorkflowThe git-workflow skill documents best practices for Git version control, branching strategies, and collaborative development. It compares GitHub Flow for continuous deployment, trunk-based development with feature flags for high velocity, and GitFlow for scheduled enterprise releases with develop and release branches. Conventional Commits cover feat fix docs style refactor test chore perf ci revert types with good versus bad examples and optional .gitmessage template. Merge versus rebase guidance explains when to preserve history versus linearize local branches, with explicit warnings against rebasing shared or protected branches. Pull request workflow includes title format, description template with what why how testing checklist, and reviewer and author checklists. Additional sections cover conflict resolution, branch naming conventions, stash workflow, semantic versioning tags, changelog generation, git config aliases, gitignore patterns, common workflows for features and fork sync, undo recipes, hooks, and documented anti-patterns like committing secrets or force pushing main.4.9kinstalls 361Skill ComplyThe skill-comply skill measures whether coding agents actually follow skills, rules, or agent definitions. It auto-generates expected behavioral sequences from any markdown target, creates scenarios at three prompt strictness levels from supportive to neutral to competing, runs claude -p capturing stream-json tool traces, classifies tool calls against spec steps with LLM not regex, checks temporal ordering deterministically, and produces self-contained reports with specs, prompts, compliance scores, and timelines. Supported targets include skills SKILL.md files, rules in rules/common, and agent definitions in agents. Key concept is prompt independence: measuring compliance even when prompts do not explicitly support the rule. Usage via uv run python -m scripts.run with dry-run for spec and scenarios only. Optional hook promotion recommendations appear for low-compliance steps. Activate via /skill-comply path or when verifying new rules and skills during quality maintenance.4.9kinstalls 362Click Path AuditThe click-path-audit skill ユーザー向けボタン/タッチポイントを完全な状態変更シーケンスを通して追跡し、機能が個別に機能するが互いにキャンセルされたり、間違った最終状態を生成したり、UIを矛盾した状態にしたままにするバグを見つけます。次の場合に使用します：体系的なデバッグがバグを見つけたが、ユーザーは壊れたボタンを報告する場合、または共有状態ストアに触れる主要なリファクター後。. /click-path-audit — 行動フロー監査静的コード読み取りが見落とすバグを見つけ：状態相互作用の副作用、順序を付けられた呼び出し間の競合状態、および互いに静かに取り消すハンドラー。この解決する問題従来のデバッグチェック： - 関数が存在しますか？（不足している配線） - クラッシュしますか？（ランタイムエラー） - 正しいタイプを返しますか？（データフロー）しかし、それはチェック**しません**： - **最終UI状態がボタンラベルが約束したものと一致しますか？** - **関数Bが関数Aが行ったばかりをサイレンス的に取り消しますか？** - **共有状態（Zustand/Redux/context）に意図した操作をキャンセルする副作用がありますか？** 実例：「新しいメール」ボタンが`setComposeMode(true)`を呼び出してから`selectThread(null)`。両方は個別に機能しました。しかし、`selectThread`には`composeMode: false`をリセットする副作用がありました。ボタンは何もしなかった。54のバグは体系的なデバッグによって見つかりました — これは見落とされました。 Agents should read SKILL.md quick start steps, verify required binaries and environment variables, and follow reference files for exact parameters before calling tools. The workflow includes decision gates, anti-patterns, and cross-links to sibling skills so agents stay aligned with upstream documentation rather than improvising. Use when user language matches documented trigger phrases or when the task clearly fits the capability described in the skill desc.4.9kinstalls 363Team Agent Orchestrationteam-agent-orchestration is a Claude Code skill for managing multiple coding agents like a team rather than a single assistant. It uses work items, explicit ownership, an agent Kanban board, branch isolation, and merge gates to keep parallel agent work visible and mergeable. A developer uses it when a task spans multiple agents, harnesses, branches, or worktrees and existing fan-out produces output but not mergeable product. It matters because it prevents agent soup, invisible work, and overlapping writes.4.8kinstalls 364Nestjs PatternsThe nestjs-patterns skill documents production-grade NestJS architecture for modular TypeScript backends. Activate when building APIs, structuring modules and controllers, adding DTO validation, guards, interceptors, config, and database integrations. Project layout keeps domain code in feature modules with cross-cutting filters, guards, and pipes in common, and DTOs colocated with owning modules. Bootstrap enables global ValidationPipe with whitelist and forbidNonWhitelisted, ClassSerializerInterceptor, and HttpExceptionFilter for consistent API behavior. Modules export services explicitly; controllers stay thin and delegate to providers with ParseUUIDPipe and typed DTOs on routes. Patterns cover configuration validation, Prisma or TypeORM integration, JWT auth strategies, testing units and HTTP endpoints, and exception filters for production services. Follow these conventions when scaffolding new NestJS services or reviewing existing module boundaries, dependency injection graphs, and validation configuration across public HTTP APIs in TypeScript codebases.4.8kinstalls 365Santa MethodThe santa-method skill implements multi-agent adversarial verification with a convergence loop. A generator produces the artifact, then two independent reviewers with no shared context evaluate the same rubric in parallel. Both must pass before output ships; otherwise a fix cycle collects all flags, repairs issues, and re-runs dual review until convergence or escalation. Use before public deploy, compliance-sensitive releases, customer-facing copy, or high hallucination-risk content where single-reviewer bias is unacceptable. Do not use for internal drafts or tasks already covered by build, test, and lint pipelines. Architecture phases are generate, dual independent review, verdict gate, and fix cycle with a configurable maximum iteration count. Reviewers return structured pass or fail per rubric criterion with cited issues rather than prose summaries. The method targets outputs where accuracy, brand constraints, or regulatory language require independent double-checking rather than spot review alone.4.8kinstalls 366Laravel VerificationThe laravel-verification skill Bucle de verificación para proyectos Laravel: verificaciones de entorno, linting, análisis estático, pruebas con cobertura, escaneos de seguridad y preparación para despliegue.. Bucle de Verificación Laravel Ejecutar antes de PRs, después de cambios importantes y antes del despliegue. Cuándo Usar - Antes de abrir un pull request para un proyecto Laravel - Después de refactorizaciones importantes o actualizaciones de dependencias - Verificación previa al despliegue para staging o producción - Ejecutar el pipeline completo de lint -> prueba -> seguridad -> preparación para despliegue Cómo Funciona - Ejecutar las fases secuencialmente desde las verificaciones de entorno hasta la preparación para despliegue, de modo que cada capa construya sobre la anterior. - Las Agents should read SKILL.md quick start steps, verify required binaries and environment variables, and follow reference files for exact parameters before calling tools.4.8kinstalls 367Brand VoiceBrand Voice extracts durable writing style profiles from real source material - posts, essays, launch notes, emails, docs, or site copy - then generates a structured VOICE PROFILE block for reuse across content workflows. Instead of re-deriving style from scratch or defaulting to generic AI patterns, developers and content teams capture the actual rhythms, compression, capitalization norms, parenthetical use, claim sharpness, and transitions from canonical sources. The skill gathers 5-20 representative samples, prioritizes recent original material, separates public from private voice lanes when needed, and bans generic tropes like fake curiosity hooks, forced lowercase, and LinkedIn thought-leader cadence. The output profile feeds directly into downstream skills like content-engine, crosspost, and lead-intelligence for consistent outbound across X, LinkedIn, email, and product channels. Extract rhythm, sentence length, compression, and capitalization from 5-20 real source samples Structured VOICE PROFILE output integrates into content-engine, crosspost, and outbound workflows Separates public launch voice from private working voice when source material splits clearly Hard bans on.4.7kinstalls 368Dynamic Workflow Modedynamic-workflow-mode is a Claude Code skill for designing task-local agent harnesses instead of only following a static command flow. A developer uses it when a task needs a custom loop, evaluator, watcher, or local dashboard, and wants each harness to carry an objective, eval gate, and handoff artifact. It also defines when to promote a one-off harness into a shared reusable skill and how to expose workflow state through control-pane checkpoints.4.7kinstalls 369BenchmarkThe benchmark skill measures performance baselines and detects regressions before pull requests, launches, or stack comparisons. Mode one captures browser Core Web Vitals via browser MCP: LCP under 2.5s, CLS under 0.1, INP under 200ms, FCP under 1.8s, and TTFB under 800ms plus bundle and third-party weight. Mode two benchmarks API endpoints with p50, p95, p99 latency over repeated calls and concurrent load. Mode three tracks cold build, HMR, test suite, TypeScript check, lint, and Docker build durations for developer feedback loops. Mode four stores baselines in .ecc/benchmarks JSON tracked in git and compares after changes with delta verdicts. Integrates with canary-watch for post-deploy monitoring and browser-qa for pre-ship checklists. Run when users report slowness, before launch, or when comparing stack alternatives with structured before-and-after tables showing metric deltas and warning thresholds for regressions across frontend, API, and build pipelines in continuous integration workflows.4.7kinstalls 370Healthcare Phi ComplianceThis skill provides patterns for protecting Protected Health Information (PHI) and Personally Identifiable Information (PII) in healthcare applications across HIPAA, GDPR, and DISHA jurisdictions. It covers three-layer protection: data classification (PHI vs PII identification), access control via Row-Level Security (RLS) for multi-facility isolation, and audit logging with tamper-proof insert-only policies. Developers use this when building patient record systems, APIs returning clinical data, or access control for clinical staff. Key workflows include schema tagging columns as PHI/PII, enforcing facility-level data isolation through RLS policies, logging access with opaque record IDs, and eliminating common leak vectors (error messages, console output, URL parameters, browser storage, service keys). Data classification framework distinguishing PHI (patient name, DOB, diagnoses, national IDs) from PII (clinician salary, vendor payments) Row-Level Security patterns for multi-facility isolation: doctors at Facility4.7kinstalls 371SeoThis skill systematically improves search visibility through technical correctness, performance optimization, and content relevance rather than tactics. Developers use it to audit crawlability, indexability, canonicals, and redirects; optimize title tags, meta descriptions, and heading structures; implement structured data (Article, Product, Organization schemas); improve Core Web Vitals (LCP, INP, CLS); conduct keyword research and map keywords to URLs; and plan internal linking and sitemap/robots changes. The workflow prioritizes fixing technical blockers before content optimization, ensures one primary search intent per page, and grounds all recommendations in page-specific, implementable actions following mobile-first indexing principles. Audits crawlability (robots.txt, noindex, redirect chains, canonical tags), indexability (URL consistency, hreflang, sitemaps, deduplication), and performance (Core Web Vitals targets: LCP <2.5s, INP <200ms, CLS <0.1) Implements structured data schemas: Organization/Business for homepage, Article/BlogPosting for editorial, Product with4.7kinstalls 372Healthcare Emr Patterns電子医療記録 EMR パターン相互運用性およびHL7 FHIR統合 name healthcare-emr-patterns description 電子医療記録 EMR パターン相互運用性およびHL7 FHIR統合 origin Health1 Super Speciality Hospitals contributed by Dr Keyur Patel version 1 0 0 Healthcare EMR Development Patterns Patterns for building Electronic Medical Record EMR and Electronic Health Record EHR systems Prioritizes patient safety clinical accuracy and practitioner efficiency When to Use Building patient encounter workflows complaint exam diagnosis prescription Implementing clinical note-taking structured free text voice-to-text Designing prescription medication modules with drug interaction checking Integrating Clinical Decision Support Systems CDSS Building lab result displays with reference range highlighting Implementing audit trails for clinical data Designing healthcare-accessible UIs for clinical data entry How It Works Patient Safety First Every design decision must be evaluated against Could this harm a patient Drug interactions MUST alert not silently pass Abnormal lab values MUST be visually flagged Critical vitals MUST trigger escalation workflows No clinical data modification without audit trail Single-Page Encounter Flow Clinical encounters shou.4.6kinstalls 373Laravel Plugin DiscoveryLaravel Plugin Discovery enables developers to find and assess Laravel packages through the LaraPlugins.io MCP server without requiring API keys. Developers use it when selecting dependencies for features like authentication, permissions, or admin panels, needing to verify package health, maintenance status, and Laravel/PHP version compatibility before integration. The skill provides two core workflows: searching packages by keyword with filters for health score (Healthy/Medium/Unhealthy), Laravel versions 5-13, PHP versions 7.4-8.5, and vendor reputation; and retrieving detailed package metrics including health indicators, last activity dates, version history, and risk scores. Best practices emphasize filtering by health status for production use, matching target Laravel versions, checking vendor reputation, and reviewing comprehensive details before recommending packages to teams. Search packages by keyword, health band (Healthy/Medium/Unhealthy), Laravel version, PHP version, and vendor name via SearchPluginTool4.6kinstalls 374Kubernetes Patternskubernetes-patterns is a Claude Code skill that provides production-grade Kubernetes workload patterns and kubectl debugging commands. A developer uses it when writing manifests, configuring resource limits and probes, setting up RBAC, or debugging CrashLoopBackOff, OOMKilled, and pending pods. It supplies copy-pasteable YAML templates with security contexts, autoscaling, and an anti-patterns checklist.4.6kinstalls 375Repo Scanrepo-scan classifies source files across C++, Java/Android, iOS (OC/Swift), and Web (TS/JS/Vue) ecosystems in a single pass, identifying which code is original, third-party, or build artifact. It detects over 50 known libraries (FFmpeg, Boost, OpenSSL) with version extraction and assigns four-stage verdicts (core asset, extract-integrate, rebuild, retire) to each module. Developers use it when inheriting large legacy codebases, preparing major refactors, auditing undeclared source-embedded dependencies, or reorganizing monorepos. It generates interactive dark-theme HTML reports with hierarchical drill-down navigation and supports multi-level scanning for monorepo summaries plus sub-project analysis. Cross-stack file classification: C++, Java/Android, iOS, and Web in one scan Library detection: identifies 50+ known libraries and extracts version numbers Four-stage module verdict system: core asset, extract-integrate, rebuild, retire Adjustable depth levels: fast, standard, deep, full - scales from 1-2 to4.6kinstalls 376Safety GuardSafety Guard intercepts destructive commands and file operations before execution in Claude Code environments. It operates in three modes: Careful mode warns on risky patterns (rm -rf, git push --force, DROP TABLE, docker system prune, kubectl delete, chmod 777, sudo rm, npm publish); Freeze mode locks write access to specified directory trees while allowing reads elsewhere; Guard mode combines both protections for maximum safety during autonomous agent execution. Developers use this when agents run full-auto, managing production systems, performing migrations, or handling sensitive data changes. Blocked actions log to ~/.claude/safety-guard.log. Integrates via PreToolUse hooks on Bash, Write, Edit, and MultiEdit tools. Careful mode detects 11+ destructive patterns before execution with suggestions for safer alternatives Freeze mode restricts write access to specified directories while preserving global read capability Guard mode combines4.6kinstalls 377CouncilGet structured, multi-perspective advice when facing ambiguous product or technical decisions with no clear winner.4.6kinstalls 378Autonomous Agent HarnessThis skill converts Claude Code into a fully autonomous agent framework using native crons, persistent memory (built-in and MCP-backed), dispatch for remote agents, and computer use capabilities. Developers use it to replace standalone frameworks like Hermes or AutoGPT with Claude Code's native infrastructure. Key workflows include setting up scheduled recurring tasks via cron expressions, storing cross-session context in markdown memory files or MCP knowledge graphs, and combining scheduled execution with computer-use automation. The system emphasizes safety boundaries - explicit user consent, local dry-run planning, and credential exclusion from reusable artifacts. Activation occurs when users request continuous agents, periodic automation, personal AI assistants with session memory, or want to replicate self-directed agent loops. Persistent memory system combining Claude Code built-in storage (~/.claude/projects/*/memory/) and MCP knowledge graphs with queryable entity-relation structures Native4.6kinstalls 379Token Budget AdvisorToken Budget Advisor (TBA) intercepts Claude's response flow before answering, allowing users to select answer depth (25%, 50%, 75%, 100%) based on token budget constraints. It estimates input tokens using prose (words × 1.3) or code (chars / 4) heuristics, classifies prompt complexity (simple to complex), applies multiplier ranges (3× - 40×) to estimate full response windows, then presents depth options with token estimates. Users trigger it by mentioning token count, response length, or depth control; shortcut phrases like "TL;DR" or "detailed" skip the menu. Each depth level defines content scope: 25% gives core answer only; 50% adds context and one example; 75% includes alternatives and analysis; 100% covers everything. Accuracy is ~85 - 90% (±15%) using heuristic estimation, not tokenization. Estimates input tokens via prose (×1.3) or code (÷4) heuristics before responding Classifies complexity (simple - complex) and applies 3× - 40× multiplier ranges to predict response windows Presents four depth levels with estimated token counts; users select 25%, 50%, 75%, or 100% Supports shortcut triggers ("short", "TL;DR", "detailed")4.6kinstalls 380Tastetaste is a Claude Code skill that adds a creative-direction layer for music videos and short-form edits in the angelcore / cloud-trance / hyperpop visual family. It distills a named-genre aesthetic vocabulary, a mood plus color plus light system, and a beat-synced editing grammar. A developer uses it to pick a coherent visual direction before generating AI b-roll and to chain ECC video skills into one production pipeline. It matters because it decides what frames should look like and in what order, so the result reads as one intentional thing.4.6kinstalls 381Product LensPressure-test product ideas, clarify user pain, and produce a go/no-go brief before turning a vague request into code.4.6kinstalls 382Intent Driven Developmentintent-driven-development is a Claude Code skill that produces verifiable acceptance criteria for ambiguous or high-impact changes. A developer uses it to clarify a feature, de-risk a security or migration change, or prepare a handoff artifact for another agent before or alongside implementation. It inspects repo context first, asks only unavoidable questions, and writes observable criteria instead of vague requirements.4.6kinstalls 383Dart Flutter PatternsThe dart-flutter-patterns skill supplies production-ready Dart and Flutter idioms for null safety, immutable state, async composition, widget architecture, and popular state management choices. Use it when starting features needing BLoC, Riverpod, or Provider patterns, reviewing null-safe Dart, wiring GoRouter navigation, implementing Dio networking, or testing widgets and cubits. Patterns cover avoiding bang operators, preferring sealed AsyncState variants, concurrent Future.wait usage, and safe BuildContext after await. Widget guidance extracts classes instead of methods, propagates const, and scopes rebuilds narrowly. State sections document BLoC events, Riverpod notifiers, and derived providers. Navigation examples wire GoRouter refreshListenable auth guards with redirect rules. Networking shows Dio interceptors and one-time token refresh retry guards. Error handling covers global capture, ErrorWidget.builder, and crash reporting hooks. Testing notes include bloc_test units and ProviderScope widget overrides with fakes over mocks.4.6kinstalls 384Agent Payment X402agent-payment-x402 adds HTTP 402 payment execution to AI agents with policy-gated spending controls and non-custodial wallet management. Supports Base via agentwallet-sdk and X Layer via OKX Payments protocol. Agents negotiate prices, check budgets, and sign transactions only within orchestrator-defined policies. Spending enforcement includes per-task budgets, session limits, allowlisted recipients, and rate limits. Exposes MCP tools (get_balance, send_payment, check_spending, list_transactions) for Claude Code and agent harnesses. Used when agents must purchase API access, charge other services, enforce spend limits, or manage ERC-4337 smart account wallets without custodial risk. x402 protocol extends HTTP 402 to machine-negotiable payment flows with budget confirmation before signing Spending policies enforced per-task, per-session, with recipient allowlists and rate limits set by orchestrator before delegation Non-custodial ERC-4337 smart accounts - agents hold own keys, no pooled funds or custodial risk MCP tools (get_balance, send_payment, check_spending, list_transactions) integrate with Claude Code and agent harnesses Supports Base Sepolia (dev), Base mainnet (prod) via ag.4.6kinstalls 385Codehealth Mcpcodehealth-mcp is a Claude Code skill that connects the CodeScene Code Health MCP server to give design-level maintainability scores. It reviews a file before edits, checks the score delta after changes, and gates commits and pull requests against Code Health regressions. A developer uses it as a structural safeguard alongside lint and tests when refactoring or after large agent-written diffs. It runs the @codescene/codehealth-mcp server over stdio via npx and needs a CS_ACCESS_TOKEN.4.5kinstalls 386Canary WatchCanary-watch monitors deployed URLs for regressions by checking HTTP status codes, console errors, network failures, Core Web Vitals (LCP/CLS/INP), critical content elements, and API endpoint SLAs. Developers use it post-deployment, after merging risky PRs, or following dependency upgrades. It runs in quick-check mode (single pass), continuous monitoring (N minutes over M hours), or diff mode comparing staging vs production. Critical thresholds trigger desktop notifications and optional Slack/Discord webhooks. Integrates with /benchmark for performance comparison and /browser-qa for full UI testing. Monitors 6 regression categories: HTTP status, console errors, network failures, Core Web Vitals, content integrity, API health Three monitoring modes: quick-check (single pass), continuous (interval-based over duration), diff (staging vs production comparison) Tiered alerting - critical (HTTP != 200, 5+ new errors, LCP > 4s), warning (LCP +500ms, CLS > during launch windows.4.5kinstalls 387CkContext Keeper (ck) is a Claude Code skill that provides persistent, per-project memory using native storage and automatic session recovery. Developers invoke commands like /ck:init to register projects, /ck:save to capture session state (summary, decisions, blockers, next steps), and /ck:resume to reload full project briefs with git-tracked changes. The skill auto-injects ~100 tokens of context on session start via a SessionStart hook, detects unsaved work and goal drift, and runs deterministic Node.js scripts stored locally. Commands output structured JSON that guides analysis of conversation state without requiring external APIs, making it ideal for multi-session coding projects where context loss causes rework. Auto-loads project context on session start via SessionStart hook with git activity detection Captures session state through /ck:save: summary, decisions, blockers, next steps in structured JSON Deterministic Node.js scripts4.5kinstalls 388Orch Build Mvporch-build-mvp is a Claude Code skill that orchestrates bootstrapping a working MVP from a design or spec document. A developer points it at an SDD or PRD, and it extracts scope and locked decisions, orders the work into thin vertical slices, scaffolds the first end-to-end slice, then TDD-implements, reviews, and commits behind two human gates. It is a thin wrapper over the shared orch-pipeline engine and reuses the GAN generator-to-evaluator harness.4.5kinstalls 389Manim VideoThe manim-video skill Build reusable Manim explainers for technical concepts, graphs, system diagrams, and product walkthroughs, then hand off to the wider ECC video stack if needed. Use when the user wants a clean animated explainer rather than a generic talking-head script. Use Manim for technical explainers where motion, structure, and clarity matter more than photorealism. - the user wants a technical explainer animation - the concept is a graph, workflow, architecture, metric progression, or system diagram - the user wants a short product or launch explainer for X or a landing page - the visual should feel precise instead of generically cinematic - manim CLI for scene rendering - ffmpeg for post-processing if needed - video-editing for final assembly or polish - remotion-video-creation when the final package needs composited UI, captions, or additional motion layers - short 16:9 MP4 - one thumbnail or poster frame - storyboard plus scene plan4.5kinstalls 390Orch Pipelineorch-pipeline is a Claude Code skill that defines the shared orchestration engine for the orch-* skill family. It specifies the gated Research-Plan-TDD-Review-Commit pipeline, a size classifier that scales ceremony to blast radius, the agent and command map for each phase, and the two human gates. A developer reads it directly only when adding a new orch operation or tuning the shared phases, gates, or agent map.4.5kinstalls 391Inherit Legacy Styleinherit-legacy-style is a Claude Code skill that prevents AI code-style drift in legacy projects. A developer uses it when onboarding an AI coding agent onto a hand-written codebase, so the model does not impose pretrained mainstream idioms. It scans the codebase across file anatomy, state and control flow, infrastructure, and error handling, resolves conflicts one at a time, and writes an enforceable .ai-style-rules.md.4.5kinstalls 392Orch Add Featureorch-add-feature is a Claude Code skill that orchestrates building a net-new feature end to end. A developer uses it when adding a capability that does not exist yet, delegating research, planning, TDD implementation, review, and gated commit to the matching ECC agents. It is a thin wrapper over the shared orch-pipeline engine and stops at a plan-approval gate and a pre-commit gate.4.5kinstalls 393Ui DemoThe ui-demo skill Playwrightを使用して美しいUIデモ動画を録画する。ユーザーがWebアプリのデモ、ウォークスルー、スクリーン録画、またはチュートリアル動画の作成を求める場合に使用する。可視カーソル、自然なリズム、プロフェッショナルな仕上がりのWebM動画を生成する。. Playwrightの動画録画機能を使用して、注入されたカーソルオーバーレイ、自然なリズム、ナラティブフローを備えた美しいWebアプリのデモ動画を録画する。ユーザーが「デモ動画」「スクリーン録画」「操作デモ」または「チュートリアル」を求める場合ユーザーが機能またはワークフローを視覚的に見せたい場合ユーザーがドキュメント、オンボーディング、ステークホルダーへのデモのために動画が必要な場合すべてのデモは探索 - リハーサル - 録画の3つのフェーズを経る。録画フェーズに直接ジャンプしない。見たことのない内容のスクリプトは書けない。フィールドが <textarea ではなく <input の場合、ドロップダウンが <select ではなくカスタムコンポーネントの場合、コメントボックスが @mentions や tags をサポートしている場合があある。仮定は録画を静かに壊す。見たことのない内容のスクリプトは書けない。フィールドが <textarea ではなく <input の場合、ドロップダウンが <select ではなくカスタムコンポーネントの場合、コメントボックスが @mentions や tags をサポートしている場合があある。仮定は録画を静かに壊す。 javascript // Run this for each page in the flow BEFORE writing the demo script const fields = await page.evaluate(() = { const els = []; document.querySelectorAll('input, select, textarea, button, [contenteditable]').forEach(el = { if (el.offsetParent !== null) { els.push({ tag: el.tagName, type: el.type '', name: el.name '', placeholder: el.placeholder '', text: el.textContent?.trim().substring(0, 40) '', contentEditable: el.contentEditable === 'true', role: el.getAttribute('role') '', }); } }); ret.4.5kinstalls 394Orch Fix Defectorch-fix-defect is a Claude Code skill that orchestrates fixing a bug end to end. A developer uses it when existing behavior is broken or wrong, first reproducing the bug as a new failing regression test, then fixing to green, reviewing, and committing behind a gate. It is a thin wrapper over the shared orch-pipeline engine and can scope unclear root causes with code-explorer.4.5kinstalls 395Orch Refine Codeorch-refine-code is a Claude Code skill that orchestrates a behavior-preserving refactor. A developer uses it when the structure should improve but behavior must not change, confirming the relevant tests are green first, then restructuring in small steps while keeping tests green. It is a thin wrapper over the shared orch-pipeline engine and can delegate dead-code and duplication sweeps to the refactor-cleaner agent.4.5kinstalls 396Healthcare Cdss PatternsThe healthcare-cdss-patterns skill 臨床意思決定支援システム（CDSS）パターン、医学的推論、およびエビデンスベースの実装。. Patterns for building Clinical Decision Support Systems that integrate into EMR workflows. CDSS modules are patient safety critical zero tolerance for false negatives. - Implementing drug interaction checking - Building dose validation engines - Implementing clinical scoring systems (NEWS2, qSOFA, APACHE, GCS) - Designing alert systems for abnormal clinical values - Building medication order entry with safety checks - Integrating lab result interpretation with clinical context The CDSS engine is a pure function library with zero side effects. Input clinical data, output alerts. This makes it fully testable. 1. checkInteractions(newDrug, currentMeds, allergies) Checks a new drug against current medications and known allergies. Returns severity-sorted InteractionAlert[]. Uses DrugInteractionPair data model. 2. validateDose(drug, dose, r 1. checkInteractions(newDrug, currentMeds, allergies) — Checks a new drug against current medications and known allergies. Returns severity-sorted InteractionAlert[]. Uses DrugInteractionPair data model. 2. validateDose(drug, dose, route, weight, age, renalFunction) — Va.4.5kinstalls 397Orch Change Featureorch-change-feature is a Claude Code skill that orchestrates altering an existing, working feature to new desired behavior. A developer uses it when a feature is not broken but should behave differently, updating the existing tests to the new spec first and then changing the implementation to match. It is a thin wrapper over the shared orch-pipeline engine and commits only after plan and pre-commit gates.4.5kinstalls 398Healthcare Eval HarnessThe healthcare-eval-harness skill ヘルスケアAIモデル評価ハーネス、臨床メトリクス、およびレギュレーション遵守の検証。. Automated verification system for healthcare application deployments. A single CRITICAL failure blocks deployment. Patient safety is non-negotiable. Note: Examples use Jest as the reference test runner. Adapt commands for your framework (Vitest, pytest, PHPUnit, etc.) the test categories and pass thresholds are framework-agnostic. - Before any deployment of EMR/EHR applications - After modifying CDSS logic (drug interactions, dose validation, scoring) - After changing database schemas that touch patient data - After modifying authentication or access control - During CI/CD pipeline configuration for healthcare apps - After resolving merge conflicts in clinical modules The eval harness runs five test categories in order. The first three (CDSS Accuracy, PHI Exposure, Data Integrity) are CRITICAL gates requiring 100% pass rate a single f The eval harness runs five test categories in order. The first three (CDSS Accuracy, PHI Exposure, Data Integrity) are CRITICAL gates requiring 100% pass rate — a single failure blocks deployment. The remaining two (Clinical Workflow, Integration) are HIGH gates requiring 9.4.5kinstalls 399Dashboard BuilderDashboard Builder is a skill for constructing actionable monitoring dashboards on platforms like Grafana and SigNoz. Developers use it when they need to translate a raw metrics list into a structured dashboard that answers operator questions: is the system healthy, where are the bottlenecks, what changed, and what action is required. The workflow enforces starting from operational questions organized around health/availability, latency/performance, throughput/volume, and resource saturation before touching any visual layout. It requires inspecting the target platform's existing dashboard JSON structure and query language before selecting only the metrics operators actually alert on and respond to. Panels are grouped by question, and each must include a title, units, and reasonable thresholds before delivery. Workflow begins with operator questions - not visual layout - organized across health, latency, throughput, saturation, and service-specific risk4.5kinstalls 400Brand Discoverybrand-discovery is a Claude Code skill that conducts a structured multi-session interview to articulate a brand's identity. It works through eight modules (purpose, positioning, audience, personality, voice, narrative, founder tension, synthesis) using laddering, 5 Whys and projective techniques. A founder or strategist uses it to produce a master brandbook. It persists session state to disk so interviews can span days and resume where they left off.4.5kinstalls 401Vue Patternsvue-patterns is a Claude Code skill that encodes Vue.js 3 Composition API patterns and best practices. It covers component architecture, reactivity, Pinia state management, Vue Router navigation, composables, testing, and Nuxt SSR patterns. A developer uses it when a project uses Vue, Nuxt, Vite plus Vue, or Pinia, or when reviewing Vue single-file components. It matters because it standardizes how Vue components, stores, and composables are structured across a codebase.4.5kinstalls 402Customer Billing OpsThe customer-billing-ops skill Stripeなどの接続された請求ツールを使用して、サブスクリプション、返金、チャーントリアージ、請求ポータルの回復、プラン分析などの顧客請求ワークフローを操作します。顧客を助けたい、サブスクリプション状態を検査したい、または収益に影響する請求操作を管理したい場合に使用します。. このスキルは、汎用的な支払いAPI設計ではなく、実際の顧客業務のために使用します。目標は、オペレーターが以下に答えるのを助けることです：この顧客は誰か、何が起きたか、最も安全な修正は何か、どのようなフォローアップを送るべきか？ - 顧客が請求が壊れている、返金が必要、またはキャンセルできないと言っている場合 - 重複サブスクリプション、偶発的な請求、失敗した更新、またはチャーンリスクを調査する場合 - プランミックス、アクティブなサブスクリプション、年間対月間の変換、またはチームシートの混乱をレビューする場合 - 請求ポータルフローの作成または検証 - サブスクリプション、請求書、返金、または支払い方法に関するサポートの苦情を監査する場合 - Stripeなどの接続された請求ツールを最初に使用する - メール、GitHub、または課題トラッカーは補足証拠としてのみ使用する - プラットフォームが必要なコントロールをすでに提供している場合、カスタムアカウント管理コードよりもホストされた請求/顧客ポータルを優先する - Stripeなどの接続された請求ツールを最初に使用する - メール、GitHub、または課題トラッカーは補足証拠としてのみ使用する - プラットフォームが必要なコントロールをすでに提供している場合、カスタムアカウント管理コードよりもホストされた請求/顧客ポータルを優先する - レスポンスに秘密鍵、完全なカード詳細、または不必要な顧客PIIを公開しない - 盲目的に返金しない。まず問題を分類する - 以下を区別する： - 偶発的な重複購入 - 意図的なマルチシートまたはチーム購入 - 壊れた製品/未達成の価値 - 失敗または不完全なチェックアウト - セルフサーブコントロールの欠如によるキャンセル - 年間プラン、チームプラン、按分状態については、アクションを取る前に契約の形状を確認する - 顧客メール - StripeカスタマーID - サブスクリプションID - 請求書ID - GitHubユーザー名または請求に紐づくことがわかっているサポートメール - 顧客 - アクティブなサブスクリプション - キャンセルされたサブスクリプション - 請求書 - 重複するアクティブなサブスクリプションなどの明らかな異常ケース典型的なアクション ------ ---------------- 重複する個人サブスクリ.4.5kinstalls 403Agent Self Evaluationagent-self-evaluation is a Claude Code skill that makes an agent rate its own output after a non-trivial task. It scores five axes (accuracy, completeness, clarity, actionability, conciseness) on a 1-5 scale with concrete evidence per axis, then lists ranked improvements. A developer uses it as a reflection step to catch omissions and overconfidence before the user does. It ships an evaluation script, report template and examples.4.4kinstalls 404Config Gcconfig-gc is a Claude Code skill that garbage-collects a Claude Code configuration. It periodically scans ~/.claude for redundant, stale or orphaned skills, memory, hooks, permissions, MCP servers and caches, then walks the user through a confirm-each-deletion cleanup. A developer uses it to slim a bloated setup and speed up session startup. It soft-deletes first, keeps an undo path, and logs every run to gc_log.md.4.4kinstalls 405Security Bounty HunterSecurity Bounty Hunter guides developers through finding exploitable, bounty-eligible vulnerabilities in a codebase rather than performing broad best-practices reviews. It targets remotely reachable attack paths including SSRF (CWE-918), auth bypass (CWE-287), RCE via deserialization (CWE-502), SQL injection (CWE-89), command injection (CWE-78), path traversal (CWE-22), and auto-triggered XSS (CWE-79). The workflow starts with scope confirmation against program rules and SECURITY.md, then maps real entry points such as HTTP handlers, upload endpoints, webhooks, and parsers, runs semgrep as a triage input, traces code paths end-to-end, and proves user-controlled data reaches a meaningful sink. Developers produce a minimal safe PoC and verify the finding is not a duplicate before writing a structured report. Low-signal patterns like local-only pickle.loads, CLI-only eval, and missing security headers are explicitly excluded. Seven in-scope CWE patterns mapped to typical impact: SSRF, auth bypass, RCE, SQLi, command injection, path traversal, XSS4.4kinstalls 406Openclaw Persona Forgeopenclaw-persona-forge is a skill for constructing full OpenClaw AI agent persona specifications from scratch. It operates in three modes: guided interview, random gacha (calling gacha.py which draws from 8 million combinations), and polish mode for existing drafts. Each run produces identity tension analysis, character boundary rules expressed in the agent's own voice, three name candidates with rationale, and a complete avatar image generation prompt. Output is structured as SOUL.md and IDENTITY.md files written to a user-specified directory. Avatar image generation is optionally delegated to a separately installed and reviewed image-generation skill; if unavailable, the skill outputs a copyable prompt for Gemini, ChatGPT, or Midjourney. Error handling degrades gracefully at each step rather than halting. The skill targets developers building OpenClaw agents who need a repeatable, opinionated process for giving agent personas durable character rather than generic tool behavior.4.4kinstalls 407Benchmark Methodologybenchmark-methodology is a Claude Code skill that scores a scoped competitor set on nine weighted dimensions using explicit 1-5 rubrics. It produces comparable, evidence-anchored profile cards and a strategic-tension plot rather than gut-feel rankings. A developer or strategist uses it as the middle step of a competitive analysis pipeline, after scoping and before assembling the report.4.4kinstalls 408Lead IntelligenceLead Intelligence is an agent-powered pipeline that replaces tools like Apollo, Clay, and ZoomInfo by orchestrating multi-stage lead discovery and outreach. Developers activate it when a user wants to find prospects in a target vertical, build an outreach list for sales or fundraising, or map warm introduction paths through a social graph. The pipeline runs five sequential stages: signal scoring via Exa and X API, mutual ranking using a weighted bridge model, warm path discovery across direct/portfolio/alumni/event chains, enrichment from Exa/X/GitHub/LinkedIn, and personalized outreach drafting per channel. Required tooling is Exa MCP and X API; optional tools include LinkedIn browser control, Apollo API, GitHub MCP, and Apple Mail for draft creation. Outreach copy is grounded in source-derived voice profiles from the brand-voice skill rather than generic templates, with strict anti-patterns around bulk messaging, fake familiarity, and platform-agnostic copy reuse.4.4kinstalls 409Dotnet PatternsThe dotnet-patterns skill C と NET言語固有のパターン規約依存性注入 async await およびロバストで保守可能な NETアプリケーション構築のためのベストプラクティス NET Development Patterns Idiomatic Cand NET patterns for building robust performant and maintainable applications When to Activate Writing new Ccode Reviewing Ccode Refactoring existing NET applications Designing service architectures with ASP NET Core Core Principles 1 Prefer Immutability Use records and init-only properties for data models Mutability should be an explicit justified choice 2 Explicit Over Implicit Be clear about nullability access modifiers and intent 3 Depend on Abstractions Use interfaces for service boundaries Register via DI container Async Await Patterns Proper Async Usage Parallel Async Operations Options Pattern Bind configuration sections to strongly-typed objects Result Pattern Return explicit success failure instead of throwing for expected failures Repository Pattern with EF Core Middleware and Pipeline Minimal API Patterns Guard Cl Developers invoke it when tasks match the documented triggers following reference files for step-by-step execution guardrails and verification checkpoints The skill keeps recommendations tied to SKILL md evidence rather.4.4kinstalls 410Google Workspace OpsThe google-workspace-ops skill Google Workspace API操作 Sheets自動化 Gmail統合およびドキュメント管理 Google Workspace Ops This skill is for operating shared docs spreadsheets and decks as working systems not just editing one file in isolation When to Use User needs to find a doc sheet or deck and update it in place Consolidating plans trackers notes or customer lists stored in Google Drive Cleaning or restructuring a shared spreadsheet Importing repairing or reformatting a Google Slides deck Producing summaries from Docs Sheets or Slides for decision-making Preferred Tool Surface Use Google Drive as the entry point then switch to the right specialist Google Docs for text-heavy docs Google Sheets for tabular work formulas and charts Google Slides for decks imports template migration and cleanup Do not guess structure from filenames alone Inspect first Workflow 1 Find the asset Start with the Drive search surface to locate the exact file sibling Developers invoke it when tasks match the documented triggers following reference files for step-by-step execution guardrails and verification checkpoints The skill keeps recommendations4.4kinstalls 411Finance Billing OpsThe finance-billing-ops skill ECCの証拠優先の収益、価格設定、返金、チーム請求、請求モデルの実態確認ワークフロー。ユーザーが販売スナップショット、価格比較、重複請求の診断、または汎用的な支払いアドバイスではなくコードに裏付けられた請求の実態を必要とする場合に使用します。 Finance Billing Ops（財務請求業務）ユーザーが金銭、価格設定、返金、チームシート論理、またはウェブサイトや販売コピーが示唆する方法で製品が実際に動作しているかどうかを理解したい場合に使用します。これは`customer-billing-ops`より広い範囲をカバーします。そのスキルは顧客の救済措置向けです。このスキルはオペレーターの実態向けです: 収益状態、価格決定、チーム請求、コードに裏付けられた請求動作。スキルスタック関連する場合、次のECCネイティブスキルをワークフローに引き込みます: - `customer-billing-ops` 顧客固有の救済措置とフォローアップ用 - `research-ops` 競合他社の価格設定や現在の市場エビデンスが重要な場合 - `market-research` 答えが価格推奨で終わる場合 - `github-ops` 請求の実態がコード、バックログ、または関連リポジトリのリリース状態に依存する場合 - `verification-loop` 答えがチェックアウト、シート処理、エンタイトルメント動作の証明に依存する場合使用するタイミング - ユーザーがStripeの売上、返金、MRR、または最近の顧客活動を尋ねる場合 - ユーザーがチーム請求、シートごとの課金、またはクォータスタッキングがコードで実際に存在するか確認したい場合 - ユーザーが競合他社の価格比較や価格モデルのベンチマークを必要とする場合 - 質問が収益の事実と製品実装の実態を混在させる場合ガードレール - ライブデータと保存されたスナップショットを区別する - 以下を分離する: - 収益の事実 - 顧客への影響 - コードに裏付けられた製品の実態 - 推奨事項 - 実際のエンタイトルメントパスがそれを適用していない限り「シートごと」と言わない - 重複したサブスクリプションが重複した価値を意味すると仮定しないワークフロー 1. 最新の請求エビデンスから開始するライブ請求データを優先します。データがライブでない場合は、スナップショッ Developers invoke it when tasks match the documented triggers, following reference files for step-by-step execution, guardrails, and verification.4.4kinstalls 412Competitive Platform Analysiscompetitive-platform-analysis is a Claude Code skill that decides who counts as a competitor and where to find them before any benchmarking begins. It sorts candidates along generic axes and groups them into Direct, Adjacent and Aspirational tiers, producing a defensible, pruned set rather than an exhaustive census. A strategist uses it as the first step of a three-skill competitive pipeline, before benchmark-methodology.4.4kinstalls 413Gan Style HarnessThe gan-style-harness skill GAN 生成的敵対ネットワークスタイルの評価ハーネス画像生成パターンおよび品質メトリクス GAN-Style Harness Skill Inspired by Anthropic's Harness Design for Long-Running Application Development https www anthropic com engineering harness-design-long-running-apps March 24 2026 A multi-agent harness that separates generation from evaluation creating an adversarial feedback loop that drives quality far beyond what a single agent can achieve Core Insight When asked to evaluate their own work agents are pathological optimists they praise mediocre output and talk themselves out of legitimate issues But engineering a separate evaluator to be ruthlessly strict is far more tractable than teaching a generator to self-critique This is the same dynamic as GANs Generative Adversarial Networks the Generator produces the Evaluator critiques and that feedback drives the next iteration When to Use Building complete applications from a one-line prompt F Developers invoke it when tasks match the documented triggers following reference files for step-by-step execution guardrails and verification checkpoints The skill keeps recommendations tied to SKILL md evidence rather than generic advice making it suitable for.4.4kinstalls 414AccessibilityThis skill ensures digital interfaces meet WCAG 2.2 Level AA accessibility standards by implementing the POUR principle (Perceivable, Operable, Understandable, Robust) across web, iOS, and Android platforms. Developers use it to design semantic UI component specifications, audit existing code for accessibility barriers and compliance gaps, and implement new WCAG 2.2 criteria including target size minimums and focus appearance standards. The workflow maps high-level design requirements to technical attributes (ARIA roles, traits, hints on web; accessibility traits and labels on native platforms) while ensuring keyboard navigation, screen reader compatibility, focus management, and proper labeling. Key implementations include semantic HTML/native elements over custom roles, 24x24px minimum target sizes, 4.5:1 text contrast ratios, visible focus indicators, and focus containment in modals. WCAG 2.2 Level AA compliance across web (HTML/ARIA), iOS (SwiftUI traits), and Android (Compose semantics).4.3kinstalls 415Agent Introspection DebuggingThe agent-introspection-debugging skill Structured self-debugging workflow for AI agent failures using capture diagnosis contained recovery and introspection reports Agent Introspection Debugging Use this skill when an agent run is failing repeatedly consuming tokens without progress looping on the same tools or drifting away from the intended task This is a workflow skill not a hidden runtime It teaches the agent to debug itself systematically before escalating to a human When to Activate Maximum tool call loop-limit failures Repeated retries with no forward progress Context growth or prompt drift that starts degrading output quality File-system or environment state mismatch between expectation and reality Tool failures that are likely recoverable with diagnosis and a smaller corrective action Scope Boundaries Activate this skill for capturing failure state before retrying blindly diagnosing common agent-specific failure patterns applying contained recovery actions producing a structured human-readable debug report Do not Developers invoke it when tasks match the documented triggers following reference files for step-by-step execution guardrails and verification checkpoints The s.4.3kinstalls 416Opensource PipelineThe opensource-pipeline skill オープンソースパイプライン: プライベートプロジェクトをフォーク、サニタイズし、安全な公開リリースのためにパッケージ化する。3つのエージェント（フォーカー、サニタイザー、パッケージャー）を連鎖させる。トリガー: '/opensource'、'open source this'、'make this public'、'prepare for open source'。" オープンソースパイプラインスキル 3段階のパイプラインを通じて任意のプロジェクトを安全にオープンソース化する: **フォーク**（シークレット除去）→ **サニタイズ**（クリーンな状態を確認）→ **パッケージ**（CLAUDE.md + setup.sh + README）。アクティベートするタイミング - ユーザーが「このプロジェクトをオープンソース化する」または「これを公開する」と言うとき - ユーザーがプライベートリポジトリを公開リリースのために準備したいとき - ユーザーがGitHubにプッシュする前にシークレットを除去する必要があるとき - ユーザーが`/opensource fork`、`/opensource verify`、または`/opensource package`を呼び出すときコマンド | コマンド | アクション | |---------|--------| | `/opensource fork PROJECT` | 完全なパイプライン: フォーク + サニタイズ + パッケージ | | `/opensource verify PROJECT` | 既存のリポジトリにサニタイザーを実行 | | `/opensource package PROJECT` | CLAUDE.md + setup.sh + READMEを生成 | | `/opensource list` | ステージングされたすべてのプロジェクトを表示 | | `/opensource status PROJECT` | ステージングされたプロジェクトのレポートを表示 | プロトコル /opensource fork PROJECT **完全なパイプライン - メインワークフロー。** ステップ1: パラメータを収集するプロジェクトパスを解決する。PROJECTに`/`が含まれる場合、パス（絶対または相対）として扱う。それ以外の場合: 現在の作業ディレクトリ、`$HOME/PROJECT`をチェックし、見つからない場合はユーザーに尋ねる。ユーザーに尋ねる: 1. 「ど Developers invoke it when tasks match the documented triggers, following referen.4.3kinstalls 417Connections OptimizerThe connections-optimizer skill レビュー優先の整理、フォロー/追加の推薦、ユーザーの実際の声で書かれたチャネル別ウォームアウトリーチのドラフトを通じて、ユーザーのXとLinkedInネットワークを再編成します。フォローリストを整理したい、現在の優先事項に向けて成長したい、または高品質な関係を中心にソーシャルグラフのバランスを取り直したい場合に使用します。コネクションオプティマイザーアウトバウンドを一方向の見込み客リストとして扱うのではなく、ユーザーのネットワークを再編成します。このスキルが扱うこと： - Xのフォロー整理と拡大 - LinkedInのフォローとコネクション分析 - レビュー優先の整理キュー - 追加とフォローの推薦 - ウォームパスの特定 - ユーザーの実際の声でのApple Mail、X DM、LinkedInのドラフト生成起動条件 - ユーザーがXのフォローを整理したい場合 - ユーザーがフォローまたはコネクションのバランスを取り直したい場合 - ユーザーが「ネットワークを整理したい」「フォロー解除すべき人は」「フォローすべき人は」「再接続すべき人は」と言った場合 - アウトリーチの品質がコールドリスト生成だけでなくネットワーク構造に依存する場合必要な入力以下を収集または推測します： - 現在の優先事項と進行中の作業 - ターゲットの役割、業界、地域、またはエコシステム - プラットフォーム選択：X、LinkedIn、または両方 - 操作しないリスト - モード: `light-pass`、`default`、または`aggressive` ユーザーがモードを指定しない場合は`default`を使用します。ツール要件推奨 - `x-api`：Xグラフの検査と最近のアクティビティ - `lead-intelligence`：ターゲット発見とウォームパスランキング - `social-graph-ranker`：ユーザーがより広いリードワークフローとは独立してブリッジ価値を採点したい場合 - Exa / ディープリサーチ：人物と企業のエンリッチメント - `brand-voice`：アウトバウンドのドラフト前フォールバック - LinkedInの分析とドラフト作成のためのブラウザコントロール - APIカバレッジが制限されている場合のXのブラウザコントロール - メールが適切なチャネルの場合、デスクトップ自動化によるApple MailまたはMail.appのドラフト作成安全デフォルト Developers invoke it when tasks match the documented triggers, following reference files for step-by-st.4.3kinstalls 418Workspace Surface AuditThe workspace-surface-audit skill アクティブなリポジトリ、MCPサーバー、プラグイン、コネクター、環境サーフェス、ツールのセットアップを監査し、最も価値の高いECCネイティブスキル、フック、エージェント、オペレーターワークフローを推奨する。ユーザーがClaude Codeのセットアップを支援してほしい場合や、環境で実際に何が使えるかを理解したい場合に使用する。ワークスペースサーフェス監査読み取り専用の監査スキル。「このワークスペースとマシンが現在実際に何をできるか、次に何を追加または有効化すべきか？」という質問に答えるために使用する。これはセットアップ監査プラグインに対するECCネイティブの回答である。ユーザーが明示的にフォローアップの実装を要求しない限り、ファイルを変更しない。使用する場面 * ユーザーが「Claude Codeをセットアップする」「自動化を推奨する」「どのプラグインまたはMCPを使うべきか？」または「何が足りないか？」と言う * スキル、フック、コネクターをさらにインストールする前にマシンやリポジトリを監査する * 公式マーケットプラグインとECCネイティブのカバレッジを比較する * ワークフローレイヤーの欠落を見つけるために `.env`、`.mcp.json`、プラグイン設定、または接続されたアプリのサーフェスをレビューする * 機能がスキル、フック、エージェント、MCP、または外部コネクターのどれであるべきかを決定する交渉不可能なルール * シークレット値を決して印刷しない。プロバイダー名、機能名、ファイルパス、キーまたは設定が存在するかどうかのみを表示する。 * ECCがそのサーフェスを合理的に所有できる場合は、一般的な「別のプラグインをインストール」という推奨よりECCネイティブワークフローを優先する。 * 外部プラグインをベースラインとインスピレーションとして扱い、権威ある製品境界としてではない。 * 3つのことを明確に区別する： * 現在利用可能なもの * 利用可能だがECCのカプセル化が不十分なもの * 利用不可能で新しい統合が必要なもの監査の入力質問に答えるために必要なファイルと設定のみを確認する： 1. リポジトリサーフェス * `package.json`、ロックファイル、言語マーカー、フレームワーク設定、`README.md` * `.mcp.json`、`.lsp.json`、`.claude/settings*.json`、`.codex/*` * `AGENTS. Developers invoke it when tasks match the documented triggers, following reference files for step-b.4.3kinstalls 419Api Connector BuilderAPI Connector developers guides developers to extend a codebase with new API integrations following the host repository's established patterns. Rather than treating each connector as a standalone HTTP client, this skill ensures consistency across connector layout, config schemas, authentication models, error handling, and test styles. Developers inspect 2+ existing connectors to understand house style, define only the target integration surface the repo actually needs, build native layers (config, client, mapper, connector entry point), validate against source patterns, and ensure registry wiring and tests match repository conventions. This prevents architectural drift, reduces friction when adding Jira, Slack, or custom API surfaces, and maintains codebase cohesion across multiple integrations. Inspect and map 2+ existing connectors to learn file layout, abstraction boundaries, config models, and test fixtures specific to the repo. Define only required surfaces: auth flows, key entities, read/write ops, pagination, rate limiting, and webhook/polling models.4.3kinstalls 420Competitive Report Structurecompetitive-report-structure is a Claude Code skill that assembles scored competitor profile cards into a decision-grade report. It defines the sections (executive summary, landscape map, competitor tiers, benchmarking matrix, deep dives, white-space and threats, recommendations) so the report answers who to compete with, how, and where the defensible white-space is. A strategist uses it as the final step of a three-skill competitive pipeline.4.3kinstalls 421Knowledge OpsKnowledge-ops manages multi-layer knowledge systems spanning local files, MCP memory, vector stores, and Git repositories. Developers use it to ingest, organize, deduplicate, and search persistent context across sessions and machines. Core workflows include classifying incoming knowledge by type (business decisions, active roadmaps, preferences, references), deduplicating against existing entries via semantic search, storing in appropriate layers (Claude Code memory for quick access, MCP graph for relations, GitHub/Linear for live execution truth, knowledge base repos for durable synthesis), and synchronizing across sources. The system prioritizes live workspace models where code lives in cloned repos, active context resides in GitHub/Linear/working-context files, and cross-machine durable memory lives in knowledge base storage rather than shadow workspaces. Quality gates ensure no duplicates, sensitive data removal, updated indexes, correct layer selection, and cross-references. Six-layer architecture: live execution (GitHub/Linear), Claude Code memory, MCP graph, knowledge base repo, external stores (Supabase/PostgreSQL), local context/archive.4.3kinstalls 422Terminal OpsTerminal-ops is an ECC agent skill for evidence-first repository execution when users ask to run commands, inspect git state, debug CI or build failures, apply narrow fixes, or push verified changes. It is intentionally narrower than generic coding guidance: confirm the repo path and branch, read the failure surface from logs and diffs before editing, keep fixes scoped to one primary failure, and report exact execution states such as confirmed, locally changed, locally verified, committed, pushed, or blocked. The workflow integrates related ECC skills including verification-loop for post-change checks, tdd-workflow when regression coverage is needed, security-review when keys or external input are involved, github-ops for CI and PR status, and knowledge-ops to persist verified results. Safety rules require read-only mode for audit-only requests, prefer repository-local scripts over ad hoc wrappers, never claim fixed until verification reruns, and never claim pushed until upstream confirms. Output follows a fixed template covering surface, evidence, actions, and state.4.3kinstalls 423Project Flow OpsTurn disconnected GitHub issues, PRs, and Linear tasks into one synchronized execution flow.4.3kinstalls 424Github OpsGitHub Operations provides structured workflows for managing repositories via gh CLI, covering issue triage (classification, labeling, deduplication), pull request reviews (CI checks, merge readiness, staleness detection), CI/CD debugging (failed run diagnosis, flaky test identification), release management (changelog generation, tag creation), and security monitoring (Dependabot alerts, secret scanning). Developers activate this skill when triaging issues, reviewing PRs, debugging broken CI/CD pipelines, or preparing releases. Key workflows include searching for duplicate issues, applying priority and type labels, checking PR mergeable status and review age, viewing failed workflow logs, generating changelogs from merged PRs, and reviewing dependency and secret scanning alerts. Issue triage by type (bug, feature-request, question, documentation, enhancement, duplicate, good-first-issue) and priority (critical, high, medium, low) with label automation. PR management tracking review status, CI checks, staleness (7+ days), and mergeable state via gh CLI.4.3kinstalls 425Social Graph RankerSocial-graph-ranker computes normalized weighted graph rankings to identify optimal warm introduction paths within existing Twitter/LinkedIn networks. Developers use this when they need the ranking engine itself - not full lead generation or network optimization - to score mutual followers by intrinsic value, map warm paths to target lists, and measure bridge value across first- and second-degree connections. The core model weights targets by role, industry, geography, and responsiveness, then decays bridge contributions exponentially by hop distance. It outputs tiered recommendations: tier-one warm introduction candidates with direct paths, tier-two conditional bridges, and identified graph gaps. Workflows include building weighted target sets, fetching user graphs, computing direct bridge scores, expanding second-degree candidates, and ranking by final responsiveness-adjusted score. Weighted graph ranking with configurable decay factors (default λ=0.5) for hop-distance attenuation. Two-tier bridge scoring: direct paths (full value) plus second-degree reach with discounted weight (α=0.3).4.3kinstalls 426Csharp TestingComprehensive testing patterns for .NET applications using xUnit as the primary framework, FluentAssertions for readable assertions, NSubstitute or Moq for mocking dependencies, and Testcontainers for integration tests with real infrastructure. Covers unit test structure via Arrange-Act-Assert, parameterized tests with Theory and InlineData, mocking workflows for repository and service layers, ASP.NET Core integration testing with WebApplicationFactory and in-memory databases, and database testing with Testcontainers. Includes test organization strategies, developers pattern for test data generation, and common antipatterns to avoid such as testing implementation details, shared mutable state, and poor test naming conventions. xUnit framework with Fact and Theory attributes for parameterized test cases. FluentAssertions for fluent, readable assertion syntax in test assertions. NSubstitute mocking for dependencies with verification of method calls and arguments. WebApplicationFactory integration testing for ASP.NET Core APIs with in-memory databases.4.3kinstalls 427Llm Trading Agent SecurityAdd layered security controls when giving an LLM agent authority to sign transactions or manage treasury operations.4.3kinstalls 428Hipaa ComplianceHIPAA Compliance is a specialized entrypoint for building or reviewing US healthcare systems subject to HIPAA regulations. It provides decision gates for identifying PHI (Protected Health Information), validating covered entity and business associate relationships, and enforcing data protection controls. Developers use this skill when designing patient-facing or clinician systems, evaluating third-party vendor BAA requirements, and auditing data flows. The skill chains `healthcare-phi-compliance` for concrete implementation rules (encryption, audit logging, PII classification), `healthcare-reviewer` for clinical workflow and patient safety assessment, and `security-review` for authentication and secrets hardening. Core guardrails include blocking PHI from logs, analytics, prompts, and error strings; requiring audit trails for all PHI reads/writes; enforcing minimum necessary access; and treating SaaS/LLM providers as blocked until BAA status is confirmed. Decision gates for PHI identification, covered entity classification, BAA vendor approval, and minimum necessary access scope.4.2kinstalls 429Agent Sortagent-sort builds evidence-backed install plans for Everything Claude Code (ECC) by analyzing repository structure and classifying skills, commands, rules, hooks, and extras into two buckets: DAILY (always-loaded, stack-matched) and LIBRARY (searchable, retained but not default-loaded). Developers use it when full ECC installs are too noisy or when a project needs trimmed, repeatable install decisions grounded in grep evidence rather than opinion. The workflow reads the actual repo stack (languages, frameworks, package managers, CI/hook configs), builds an evidence table linking each component to concrete file counts and config matches, then produces an install plan with verification checks to ensure stale language rules and incompatible hooks are not left active. Classifies ECC surfaces using repo-local evidence: file extensions, lockfiles, framework configs, build scripts, and dependency manifests instead of guesses. Executes parallel review passes across agents, skills, commands, rules, hooks, scripts, and extras for comprehensive component audit.4.2kinstalls 430Code TourThis skill generates CodeTour `.tour` JSON files that create guided, interactive walkthroughs of codebases. Developers use it to build onboarding tours, architecture walkthroughs, PR reviews, and root-cause analysis tours that link directly to specific files and line ranges. Each tour targets a persona (e.g., new-joiner) and tells a narrative explaining what code does, why it matters, and what to explore next. Tours live in `.tours/` directories and serve as reusable artifacts rather than ad-hoc markdown. The skill starts with codebase exploration - README, structure, config files - then constructs step-by-step sequences (9-13 steps for onboarding) without modifying source code. Generate CodeTour `.tour` JSON files with file paths, line anchors, and explanations. Target specific personas (new-joiner, reviewer, architect) with appropriate depth and narrative. Discover repo structure, entry points, and modified files before writing steps.4.2kinstalls 431Automation Audit OpsAutomation-audit-ops is an evidence-based inventory and overlap detection workflow for Everything Claude Code (ECC) environments. Operators use it to map which automations are live, which jobs are broken, where redundancy exists, and which tools perform actual work before rebuilding. The skill surfaces real automation across local hooks, GitHub Actions, MCP configs, connectors, and wrapper scripts; classifies each by live state (configured, authenticated, recently verified, stale, missing); traces evidence to concrete file paths, logs, and config entries; and returns keep/merge/cut/fix-next recommendations to consolidate noise into one canonical ECC lane. It pulls related ECC skills (workspace-surface-audit, knowledge-ops, github-ops, verification-loop) when audits need to reconcile with live repos or prove post-fix state. Maps all automation surfaces: repo hooks, GitHub Actions, MCP servers, connectors, wrapper scripts across local and CI/CD runtime. Classifies each automation by live state - configured, authenticated, recently verified, stale, or missing - with proof paths.4.2kinstalls 432Ml Adoption Playbookml-adoption-playbook is a Claude Code skill with an end-to-end methodology for adding machine learning to an existing non-ML codebase. A developer uses it to frame the problem, audit data readiness, decouple inference behind an API boundary, build a baseline model, and hand off to MLOps. It stresses starting with a heuristic, preventing data leakage, and using feature flags and fallbacks for safe rollout.4.2kinstalls 433Product CapabilityProduct Capability translates product intent (PRDs, roadmap items, discussions) into explicit engineering constraints and SRS-style capability contracts. Developers use this when implementation gaps are not about what to build, but what must be true before work starts across multiple services or teams. The skill extracts hidden assumptions, separates user-visible promises from implementation details, defines actor surfaces and state transitions, and produces a durable reusable artifact (capability manifest) that survives across sessions and harnesses. It follows a four-step workflow: restate capability precisely, resolve constraints, define implementation-facing contracts with interfaces and data implications, and translate into execution readiness. Non-negotiable rules enforce marking unresolved questions explicitly, calling out fixed policy versus open architecture decisions, and refusing to invent product truth. Converts PRD intent into one-sentence capability restatement with explicit actor, new capability, and outcome change.4.2kinstalls 434Generating Python Installergenerating-python-installer is a Claude Code skill for advanced Windows packaging of Python applications. A developer uses it to compile with Nuitka in folder (dist) mode, slim the dist folder, analyze DLL footprint, and package a residue-free installer with Inno Setup. It targets minimal installer size and fast startup, not basic script-to-exe conversion.4.2kinstalls 435Research OpsThe research-ops skill 証拠優先のECC現状調査ワークフロー。ユーザーが現在の公開証拠と提供されたローカルコンテキストに基づいて最新の事実、比較、情報の充実、または推奨事項を求める場合に使用する。. ユーザーが現在の情報の調査、オプションの比較、人物や企業情報の充実、または繰り返しのクエリをモニタリング可能なワークフローに変換することを求める場合に使用する。これはリポジトリのリサーチスタックの操作ラッパーである。deep-research、exa-search、market-research の代替品ではなく、それらをいつどのように組み合わせて使うかを指示するものである。 exa-search：現在のウェブ情報の素早い発見に使用 deep-research：引用付きの複数ソース統合に使用 market-research：最終結果が推奨または優先順位付け決定である場合に使用 lead-intelligence：タスクが一般的な調査ではなく人物/企業を対象とする場合に使用 knowledge-ops：結果を後続のコンテキスト用に永続的に保存する必要がある場合に使用ユーザーが「調査」「調べる」「比較」「誰に連絡すべきか」「最新情報」に言及する場合答えが現在の公開情報に依存する場合ユーザーが証拠を提供し、それを新しい推奨事項に組み込んでほしい場合タスクが繰り返し発生する可能性があり、一回限りのクエリではなくモニタリングに転換すべき場合ユーザーが「調査」「調べる」「比較」「誰に連絡すべきか」「最新情報」に言及する場合答えが現在の公開情報に依存する場合ユーザーが証拠を提供し、それを新しい推奨事項に組み込んでほしい場合タスクが繰り返し発生する可能性があり、一回限りのクエリではなくモニタリングに転換すべき場合新鮮な検索が安価な場合、古い記憶に頼って現在の問いに答えない以下を区別する：出典付きの事実ユーザーが提供した証拠推論推奨事項答えがローカルコードまたはドキュメントにすでに存在する場合、重い調査プロセスを起動しない素早い事実的回答比較または意思決定メモリード/情報充実処理繰り返しモニタリング候補 exa-search を使って素早い発見を行う統合または複数ソース情報が必要な場合、deep-research にアップグレードする結果を推奨形式で提示する必要がある場合、market-research を使用する実際のニーズがターゲット優先順位付けやウォームパス発見の場合、lead-intelligence に転送するユーザーが同じ調査の問いを繰り返す可能性がある場合、それを明示し、永遠に同じ手動検索を繰り返すのではなく、モニタリングまたはワークフローレイヤーの採用を提案する。 The workflow f.4.2kinstalls 436Messages OpsThe messages-ops skill 日本語翻訳：このファイルは messages-ops 用の日本語翻訳が必要です. 詳細は：D:/tmp/everything-claude-code/skills/messages-ops/SKILL.md The workflow follows documented commands, reference files, and validation steps from SKILL.md. Agents should read prerequisites, apply progressive fixes, and confirm outcomes before marking tasks complete. Security-sensitive operations require explicit user confirmation per the skill guidance. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling.4.2kinstalls 437Email OpsThe email-ops skill ECC用の証拠ベースのメールボックストリアージ、ドラフト作成、送信検証、および送信済みメールセーフフォローアップワークフロー。ユーザーがメールを整理したり、実際のメールサーフェスを通じてドラフトまたは送信したい、または送信済みメールに何が到着したかを証明したい場合に使用します。. Use this when the real task is mailbox work: triage, drafting, replying, sending, or proving a message landed in Sent. This is not a generic writing skill. It is an operator workflow around the actual mail surface. Pull these ECC-native skills into the workflow when relevant: - brand-voice before drafting anything user-facing - investor-outreach for investor, partner, or sponsor-facing mail - customer-billing-ops when the thread is a billing/support incident rather than generic correspondence - knowledge-ops when the message or thread should be captured into durable context afterward - research-ops when a reply depends on fresh external facts - brand-voice before drafting anything user-facing - investor-outreach for investor, partner, or sponsor-facing mail - customer-billing-ops when the thread is a billing/support incident rather than generic correspondence - knowledge-ops when the message or thread should be captured into durable context afterward - research-ops when a reply depends on fresh external facts - user asks t.4.2kinstalls 438Unified Notifications OpsThe unified-notifications-ops skill GitHub、Linear、デスクトップアラート、フック、接続された通信インターフェースを網羅する、統合されたECCネイティブワークフローとして通知を運用する。真の問題がアラートルーティング、重複排除、エスカレーション、またはインボックス崩壊である場合に使用する。. 真の問題が通知の欠如ではなく、通知システムの断片化にある場合にこのスキルを使用する。目標は、分散したイベントを単一のオペレーターインターフェースに統合することであり、以下を含む：明確な重大度レベル明確な責任者明確なルーティング明確な次のアクションユーザーがGitHub、Linear、ローカルフック、デスクトップアラート、チャット、メール間の統一通知チャネルを望んでいる CI失敗、レビューリクエスト、Issue更新、オペレーターイベントが各所に散在している現在のセットアップがアクションではなくノイズを生成しているユーザーが重複する通知ブランチや積み残しのプロポーザルを単一のECCネイティブチャネルに統合したいワークスペースにフック、MCP、または接続されたツールがあるが、一貫した通知戦略がないユーザーがGitHub、Linear、ローカルフック、デスクトップアラート、チャット、メール間の統一通知チャネルを望んでいる CI失敗、レビューリクエスト、Issue更新、オペレーターイベントが各所に散在している現在のセットアップがアクションではなくノイズを生成しているユーザーが重複する通知ブランチや積み残しのプロポーザルを単一のECCネイティブチャネルに統合したいワークスペースにフック、MCP、または接続されたツールがあるが、一貫した通知戦略がない GitHub Issues、PR、レビュー、コメント、CI Linear Issues/プロジェクトのステータス変更ローカルフックイベントとセッションライフサイクルシグナルデスクトップ通知プリミティブ接続されたメール/チャットインターフェース（実際に存在する場合）独立した通知製品をユーザーに勧めるより、ECCネイティブのオーケストレーションを優先する。トークン、シークレット、Webhookシークレット、内部識別子を決して公開しない以下を区別する：イベントソース重大度レベルルーティングチャネルオペレーターアクション中断コストが不明な場合はデフォルトでサマリーファーストアプローチを取るすべてのチャネルにすべてのイベントをブロードキャストしない真の解決策がより良いIssueトリアージ、フック戦略、またはプロジェクトフローである場合は明示する 1. キャプチャイベント 2. 分類緊急度と責任者 3. ルーティング適切なチャネルへ 4. マー.4.1kinstalls 439Nodejs Keccak256The nodejs-keccak256 skill JavaScriptとTypeScriptにおけるEthereumハッシュバグを防ぐ。NodeのSHA3-256はNIST SHA3であり、Ethereum Keccak-256ではなく、セレクター、署名、ストレージスロット、アドレス導出を静かに破壊する。. EthereumはKeccak-256を使用し、Nodeのcrypto.createHash('sha3-256')が公開するNIST標準化SHA3バリアントではない。 - Ethereum関数セレクターやイベントトピックの計算 - JS/TSでEIP-712、署名、Merkle、またはストレージスロットヘルパーの構築 - Nodeのcryptoを直接使用してEthereumデータをハッシュするコードのレビュー ```javascript import crypto from 'crypto'; import { keccak256, toUtf8Bytes } from 'ethers'; const data = 'hello'; const nistSha3 = crypto.createHash('sha3-256').update(data).digest('hex'); const keccak = keccak256(toUtf8Bytes(data)).slice(2); const data = 'hello'; const nistSha3 = crypto.createHash('sha3-256').update(data).digest('hex'); const keccak = keccak256(toUtf8Bytes(data)).slice(2); console.log(nistSha3 === keccak); // false typescript import { keccak256, toUtf8Bytes, solidityPackedKeccak256, id } from 'ethers'; const hash = keccak256(new Uint8Array([0x01, 0x02])); const hash2 = keccak256(toUtf8Bytes('hello')); const topic = id('Transfer(address,address,uint256)'); const packed = solidityPackedKeccak256( ['address', 'uint256'], ['0x742d35Cc6634C0532925a3b8D4C9B569890FaC1c', 100n], ); typescript import { keccak256,.4.1kinstalls 440Defi Amm SecurityThe defi-amm-security skill DeFi自動マーケットメーカー（AMM）スマートコントラクトセキュリティ監査パターン。フラッシュローン、スリッページ、サンドイッチング攻撃、価格操作、再入攻撃、不正確な整数演算をカバー。. 自動マーケットメーカー（AMM）スマートコントラクトのセキュリティパターンと監査チェックリスト。 - Uniswap、Curve、BalancerなどのようなAMMコントラクトをレビュー - スワップ関数、ミント、バーンロジックでセキュリティの問題を検出 - フラッシュローン、価格操作脆弱性を特定 - DeFiプロトコルの経済セキュリティをテスト - [ ] すべての価格参照はオラクルから取得 - [ ] フラッシュローンは考慮（価格を信頼しない） - [ ] スリッページ保護が実装 - [ ] チェック・エフェクト・相互作用パターン使用 - [ ] 再入保護（ミューテックス/チェック付き） - [ ] 整数オーバーフロー/アンダーフロー処理 - [ ] アクセス制御とロール分離 The workflow follows documented commands, reference files, and validation steps from SKILL.md. Agents should read prerequisites, apply progressive fixes, and confirm outcomes before marking tasks complete. Security-sensitive operations require explicit user confirmation per the skill guidance. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case handling. Follow SKILL.md reference sections for detailed command examples and edge-case h.4.1kinstalls 441Ecc Tools Cost AuditThe ecc-tools-cost-audit skill ECC ツール、エージェント、スキル、および実装のコスト監査を実施します。プロンプト入力トークンを分析して、計算効率を定量化します。. Use this skill when the user suspects the ECC Tools GitHub App is burning cost, over-creating PRs, bypassing usage limits, or routing free users into premium analysis paths. This is a focused operator workflow for the sibling ECC-Tools repo. It is not a generic billing skill and it is not a repo-wide code review pass. Pull these ECC-native skills into the workflow when relevant: - autonomous-loops for bounded multi-step audits that cross webhooks, queues, billing, and retries - agentic-engineering for tracing the request path into discrete, provable units - customer-billing-ops when repo behavior and customer-impact math must be separated cleanly - search-first before inventing helpers or re-implementing repo-local utilities - security-review when auth, usage gates, entitlements, or secrets are touched - verification-loop for proving reru Follow SKILL.md reference sections for detailed command examples and edge-case handling.4.1kinstalls 442Hookify RulesThe hookify-rules skill 自動フック実装、イベントドリブン実行、およびルール駆動ワークフロー。. Hookify rules are markdown files with YAML frontmatter that define patterns to watch for and messages to show when those patterns match. Rules are stored in .claude/hookify.{rule-name}.local.md files. ```markdown --- name: rule-identifier enabled: true event: bash file stop prompt all pattern: regex-pattern-here --- Message to show Claude when this rule triggers. Can include markdown formatting, warnings, suggestions, etc. ``` Field Required Values Description ------- ---------- -------- ------------- name Yes kebab-case string Unique identifier (verb-first: warn- , block- , require- ) enabled Yes true/false Toggle without deleting event Yes bash/file/stop/prompt/all Which hook event triggers this action No warn/block warn (default) shows message; block prevents operation pattern Yes regex string Pattern to match ( or use conditions for c Field Required Values Description ------- ---------- -------- ------------- name Yes kebab-case string Unique identifier (verb-first: warn- , block- , require- ) enabled Yes true/false Toggle without deleting event Yes bash/file/stop/prompt/all Which hook event triggers this action No wa.4.1kinstalls 443Evm Token DecimalsThe evm-token-decimals skill EVMチェーン全体でサイレントな小数点不一致バグを防ぐ。ランタイムでの小数点照会、チェーン対応キャッシング、ブリッジドトークンの精度ドリフト、ボット・ダッシュボード・DeFiツール向けの安全な正規化をカバーします。. サイレントな小数点不一致は、エラーを発生させることなく残高やUSD値が桁違いになる最も簡単な方法のひとつです。 - Python、TypeScript、またはSolidityでERC-20残高を読み取る場合 - オンチェーン残高から法定通貨の値を計算する場合 - 複数のEVMチェーン間でトークン量を比較する場合 - ブリッジされた資産を扱う場合 - ポートフォリオトラッカー、ボット、またはアグリゲーターを構築する場合ステーブルコインが同じ小数点を使用していると仮定しないでください。ランタイムでdecimals()を照会し、(chain_id, token_address)でキャッシュし、値の計算には小数点安全な数学を使用します。 ```python from decimal import Decimal from web3 import Web3 python from decimal import Decimal from web3 import Web3 ERC20_ABI = [ {"name": "decimals", "type": "function", "inputs": [], "outputs": [{"type": "uint8"}], "stateMutability": "view"}, {"name": "balanceOf", "type": "function", "inputs": [{"name": "account", "type": "address"}], "outputs": [{"type": "uint256"}], "stateMutability": "view"}, ] def get_token_balance(w3: Web3, token_address: str, wallet: str) - Decimal: contract = w3.eth.contract( address=Web3.to_checksum_address(token_address), abi=ERC20_ABI, ) decimals = contract.functions.decimals().call() raw = contract.functions.balanceOf(Web3.to_checksum_address(wallet)).call() return Decimal(raw) / Decimal(10 decimals).4.1kinstalls 444React Native PatternsThe react-native-patterns skill provides development patterns for building React Native mobile applications. It covers component structure, navigation, styling, state management, and platform-specific considerations for iOS and Android. Use when implementing or reviewing React Native features and you want consistent mobile development practices.4kinstalls 445Ecc RecipesECC Recipes maps workflows to the correct Everything Claude Code command groups with run order and stop conditions, filling the gap between the flat ecc-guide catalog and prompt-optimizer's single prompts. It reads the live commands directory each run, classifies orch-, multi-, and other families, and advises only—never executing tasks for the user. Trigger when asking which ECC commands run an MVP, defect fix, or refactor pipeline.3.9kinstalls 446Delivery GateDelivery Gate is an ECC Stop hook that mechanically blocks Claude Code from ending until quality checks pass. It uses regex on transcript tails for rationalization warnings, filesystem mtimes for stale learning libraries including growth-log, and disk-usage thresholds—with blocking on critical disk or multiple stale libs after complex tasks. It complements self-audit by checking machine-verifiable session hygiene, not output eloquence.3.8kinstalls 447Growth LogGrowth Log teaches how to write learning entries that compound across sessions instead of useless diary lines like fixed a bug in X. It activates after complex or painful tasks, emphasizes failures over achievements, merges duplicate root causes, and demands transferable next-time guidance. Works with Markdown, Notion, or Obsidian—skip trivial typo fixes. Pairs with ECC delivery-gate habits that enforce learning capture.3.8kinstalls 448Mailtrap Email IntegrationMailtrap Email Integration guides agents through transactional email with Mailtrap's Email API and Sandbox. It enforces separate sandbox and production endpoints, Bearer authentication, domain DNS verification before real delivery, and environment-based routing so dev tests never hit live inboxes. Includes TypeScript send examples and anti-patterns like hardcoded tokens or skipping domain verification.3.8kinstalls 449Gateguardgateguard is a PreToolUse hook that forces investigation before file edits instead of unreliable LLM self-evaluation. The three-stage gate denies the first Edit, Write, or destructive Bash attempt, forces the model to gather concrete facts, then allows retry after facts are presented. Edit and MultiEdit gates require listing all importers, affected public symbols, data schema fields and date formats with redacted values, and quoting the user instruction verbatim. Write gates demand caller file and line references, Glob confirmation no duplicate exists, and the same data schema check. Destructive Bash gates trigger on rm -rf, git reset --hard, git push --force, and drop table with rollback procedures required. Routine Bash gates once per session ask for the current request and command purpose. Documented A/B tests show gated agents averaging 9.0 versus 6.75 ungated on analytics and webhook validator tasks. Enable via scripts/hooks/gateguard-fact-force.js in hooks.json or pip install gateguard-ai with .gateguard.yml ignore paths. ECC_GATEGUARD=off disables during setup. Complements safety-guard runtime checks and post-edit code-reviewer skill.3.8kinstalls 450Plan Canvasplan-canvas helps agents create structured planning canvases that scope features and projects before implementation. It organizes goals, constraints, and milestones so teams align on what to build next. Use when starting a non-trivial effort and you want a visible plan before writing code.3.7kinstalls 451Loop Design CheckAgent loop design and review skill that helps write goal-oriented agent loops and audit them for common failure modes. Covers spinning and token burning, Goodhart-gaming verifiers, and running wrong answers to completion with machine-decidable gate patterns.3.5kinstalls 452Everything Claude Codeeverything-claude-code is an agent skill generated from affaan-m/everything-claude-code that teaches Claude the repository's JavaScript hybrid-module architecture, separate test layout, and Conventional Commits workflow. It activates on repository changes for features, tests, and documentation, enforcing fix, test, feat, and docs prefixes with imperative mood messages averaging about 65 characters. The skill exposes implicit invocation with the default prompt to follow repository conventions and workflows rather than generic coding advice. Example commit messages in the skill document feat, chore, docs, fix, and merge patterns drawn from real repository history. Developers reach for everything-claude-code when agent-assisted edits must match house style for commits, testing, and module structure inside this project. It functions as a repo-specific policy layer that keeps multi-session contributions aligned with documented patterns, commit examples, and test placement rules extracted from 500 analyzed commits in the upstream repository.3.4kinstalls 453Hermes ImportsHermes Imports converts stable Hermes operator workflows into reusable ECC skills safe for publication. Hermes is the operator shell while ECC is the reusable workflow layer; imports move patterns without moving private state. Use when a Hermes workflow has repeated enough to become reusable, a local operator prompt should become a public skill, or launch and research workflows need sanitized handoff docs. Import rules convert absolute paths to repo-relative placeholders, replace live account names with role labels, describe credentials by provider only, and block raw workspace exports, tokens, OAuth files, health data, CRM data, or finance data. The sanitization checklist scans for /Users paths, ~/.hermes paths, API keys, personal contacts, client names, revenue details, and raw private tool logs. The conversion pattern identifies the repeatable loop, strips private inputs, rewrites paths, adds When To Use sections, and runs secret scans before PR. Output contract returns candidate skill name, sanitized summary, public inputs, removed private inputs, risks, and files to create.3kinstalls 454Ito Market IntelligenceThis skill gathers prediction-market context, including events, venues, underliers, liquidity, and news, into a source-grounded brief. It is read-only, so it never places, cancels, or routes orders, and it withholds investment advice. A solo builder reaches for it when researching a market theme before building a product around prediction-market data.2.9kinstalls 455Ito Basket CompareCompares Itô prediction-market baskets against a user's knowledge base, portfolio notes, or research thesis in read-only mode. Highlights fit, exposure, assumptions, and missing context without providing investment advice. Use when inspecting basket alignment before manual trading decisions.2.9kinstalls 456Ito Data Atlas AgentThe Itô Data Atlas Agent skill guides designing background agents that watch data sources, build candidate prediction-market baskets, and draft parameter changes for human review. Solo builders use it to architect agentic research pipelines with explicit approval gates, ensuring all execution stays behind human control.2.9kinstalls 457Ito Trade PlannerBuilds structured prediction-market trade planning worksheets for Itô or venue workflows without placing trades. Inspects venues, underliers, constraints, and manual execution prerequisites in checklist form. Use when users want a planning worksheet but not investment advice or automated trading.2.9kinstalls 458Literature Review学術的または技術的な文献の本体を検索スクリーニング統合引用するタスクにこのスキルを使用しますシステマティックスコーピングまたはナラティブな文献レビューの構築研究質問に対する最先端の状態を統合するギャップ矛盾または今後の研究方向性を見つける論文やレポートの引用に裏付けられた背景セクションの準備査読済み論文プレプリント特許技術レポートにわたる証拠の比較ナラティブレビュー広範な統合方向性把握に有用スコーピングレビュー概念方法証拠のギャップをマッピングシステマティックレビュー事前定義されたプロトコル再現可能な検索明示的なスクリーニングと除外メタ分析システマティックレビューに量的効果の集計を加えたものどのレベルの厳密さが必要かをユーザーに確認してください未指定の場合探索的作業にはスコーピングレビュー公開または臨床的主張にはシステマティックレビューをデフォルトとします対象集団 Population 介入または暴露 Intervention or exposure 比較対象 Comparator アウトカム Outcome The literature review agent skill provides documented workflows prerequisites triggers and safety guidance from its SKILL md source Agents load it when user requests match the description and follow step by step instructions without inventing capabilities It integrates with standard agent tooling for the tasks inputs outputs and failure modes described in the repository documentation2.6kinstalls 459Fastapi PatternsThe fastapi-patterns skill. FastAPI patterns for async APIs, dependency injection, Pydantic request and response models, OpenAPI docs, tests, security, and production readiness. - Splitting routers, schemas, dependencies, and database access. - Writing async endpoints that call a database or external service. - Adding authentication, authorization, OpenAPI docs, tests, or deployment settings. - Checking a FastAPI PR for copy-pasteable examples and production risks. - owns Pydantic request and response models. - owns database, auth, pagination, and request-scoped dependencies. The workflow follows the source SKILL.md contract with progressive reference loading, clear trigger phrases, and practical steps developers can apply directly in agent sessions. The workflow follows the source SKILL.md contract with progressive reference loading, clear trigger phrases, and practical steps developers can apply directly in agent sessions. The workflow follows the source SKILL.md contract with progressive reference loading, clear trigger phrases, and practical steps developers can apply directly in agent sessions.2.5kinstalls 460Windows Desktop E2eE2E testing for Windows native desktop apps WPF WinForms Win32 MFC Qt using pywinauto and Windows UI Automation The windows-desktop-e2e skill documents workflows and patterns from the repository SKILL md name windows-desktop-e2e description E2E testing for Windows native desktop apps WPF WinForms Win32 MFC Qt using pywinauto and Windows UI Automation origin ECC Windows デスクトップ E2E テスト pywinauto と Windows UI Automation UIA を使用したWindowsネイティブデスクトップアプリケーションのエンドツーエンドテスト WPF WinForms Win32 MFC Qt 5 x 6 x をカバーし Qt固有のガイダンスは専用セクションとして提供しますアクティベートするタイミング Windowsネイティブデスクトップアプリケーションのエンドツーエンドテストを書くまたは実行するときデスクトップGUIテストスイートをゼロから設定するとき不安定または失敗するデスクトップオートメーションテストを診断するとき既存のアプリにテスタビリティ AutomationId アクセシブルな名前を追加するときデスクトップエンドツーエンドをCI CDパイプライン GitHub Actions windows-latest に統合するとき使用しないタイミング Webアプリケーション e2e-testing スキル Playwright を使用する Electron CEF WebView2 アプリ HTMLレイヤーにはUIAではなくブラウザオートメーションが必要モバイルアプリプラットフォーム固有のツールを使用する UIAutomator XCUITest 実行中のGUIを必要としない純粋なユニットまたは統合テストコアコンセプトすべてのWindowsデスクトップオートメーションは UI Automation UIA に依存しますこれはWindowsに組み込まれたアクセシビリティAPIですサポートされているすべてのフレームワークは読み取りおよび操作可能なプロパティを持つUIA要素のツリーを公開しますテスト Python pywinauto UIAバックエンド Windows UI Automation API Windowsに組み込みフレームワーク非依存.2.5kinstalls 461Pubmed Database生物医学文献 MeSH クエリ PMID 検索引用取得および API を利用した文献モニタリングのための PubMed および NCBI E-utilities の直接検索ワークフロー The pubmed-database skill documents workflows and patterns from the repository SKILL md name pubmed-database description 生物医学文献 MeSH クエリ PMID 検索引用取得および API を利用した文献モニタリングのための PubMed および NCBI E-utilities の直接検索ワークフロー origin community PubMed Database 一般的なウェブ検索ではなく PubMed から生物医学文献が必要なタスクにこのスキルを使用します使用するタイミング MEDLINE または生命科学文献の検索 MeSH 用語フィールドタグ日付または文献種別を使った PubMed クエリの構築 PMID アブストラクト出版メタデータまたは関連引用の検索再現可能な検索文字列が必要なシステマティックレビューの検索パスの実行 Python シェルまたは別の HTTP クライアントから直接 NCBI E-utilities を使用クエリの構築研究質問から始め概念に分割しブール演算子で概念を組み合わせます text concept_1 AND concept_2 AND filter synonym_a OR synonym_b NOT exclusion_term 有用な PubMed フィールドタグ ti タイトル ab アブストラクト tiab タイトルまたはアブストラクト au 著者 ta 雑誌タイトル略語 mh MeSH 用語 majr 主要 MeSH トピック pt 出版種別 dp 出版日 la 言語例 text diabetes mellitus mh AND treatment tiab AND systematic review pt AND 2023 2026 dp metformin nm OR insulin nm AND diabetes mellitus type 2 mh AND randomized controlled trial pt smith ja au AND cancer tiab AND 2026 dp AND english la MeSH とサブヘッディング概念が安定した統制語彙用語を持つ場合は MeSH を優先しますトピックが新しいまたは用語が多様な場合は MeSH とタイトルアブストラクト用語を組み合わせます正しいサブヘッディング構文ではサブヘッディングを.2.4kinstalls 462Mysql PatternsGet battle-tested MySQL and MariaDB patterns for schema design, query optimization, indexing, transactions, replication, and connection pools.2.4kinstalls 463Vite PatternsThe vite-patterns skill Vite build tool patterns including config, plugins, HMR, env variables, proxy setup, SSR, library mode, dependency pre-bundling, and build optimization. Activate when working with vite.config.ts, Vite plugins, or Vite-based projects. It covers vite.config.ts または vite.config.js を設定するとき. Key workflows include 環境変数または .env ファイルを設定するとき. Developers invoke vite-patterns when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from.2.4kinstalls 464Frontend Design DirectionThe frontend-design-direction skill フロントエンド設計の方向性、美的原則、および一貫した設計言語実装。 It covers the user asks to build a web page, app, dashboard, artifact, component, or UI.. Key workflows include the user asks to make an interface more polished, distinctive, beautiful, or. - The user asks to build a web page, app, dashboard, artifact, component, or UI. - The user asks to make an interface more polished, distinctive, beautiful, or less generic. - The implementation needs visual hierarchy, typography, color, motion, layout, and interaction choices. - The current UI works but reads as flat, generic, templated, or mismatched to the audience. Developers invoke frontend-design-direction when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation.2.4kinstalls 465Agent Architecture AuditThe agent-architecture-audit skill エージェントおよび LLM アプリケーション向けのフルスタック診断。12 層のエージェントスタックにおけるラッパーリグレッション、メモリ汚染、ツール規律の失敗、隠れた修復ループ、レンダリング破損を監査します。重要度順の発見事項とコードファーストの修正を生成します。エージェントアプリケーション、自律ループ、または LLM を活用した機能を構築する開発者に必須です。 It covers エージェントまたは LLM を活用したアプリケーションを本番リリースする前. Key workflows include ツール呼び出し、メモリ、または多段階ワークフローを含む機能をリリースする前. Developers invoke agent-architecture-audit when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream docume.2.4kinstalls 466Homelab Network SetupThe homelab-network-setup skill ホームラボネットワーク基盤設定、デバイス設定、接続性、およびネットワークセグメンテーション。 It covers planning a new home network or redesigning an ISP-router-only setup.. Key workflows include choosing gateway, switch, and access point roles.. - Planning a new home network or redesigning an ISP-router-only setup. - Choosing gateway, switch, and access point roles. - Designing IP ranges, DHCP scopes, static reservations, and DNS. - Preparing for future VLANs, Pi-hole, NAS, lab servers, or VPN access. - Troubleshooting a new network that has double NAT, unstable Wi-Fi, or changing server addresses. Developers invoke homelab-network-setup when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation.2.4kinstalls 467Network Interface HealthThe network-interface-health skill ルーター、スイッチ、Linuxホスト上のインターフェースエラー、ドロップ、CRC、デュプレックス不一致、フラッピング、速度ネゴシエーション問題、カウンタートレンドを診断する。 It covers ホストまたはVLANにパケットロス、レイテンシスパイク、または断続的な到達不能がある。. Key workflows include スイッチまたはルーターのインターフェースにCRC、ランツ、ジャイアント、ドロップ、リセット、またはフラップが表示されている。. Developers invoke network-interface-health when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context.2.4kinstalls 468Production AuditThe production-audit skill 日本語翻訳：このファイルは production-audit 用の日本語翻訳が必要です It covers description: 日本語翻訳：このファイルは production-audit 用の日本語翻訳が必要です. Key workflows include このファイルの翻訳は実装中です。英語版は元のスキルファイルを参照してください。. Developers invoke production-audit when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and.2.4kinstalls 469Claude ApiComprehensive guide to Anthropic Claude API patterns including Messages API, streaming, tool use, vision capabilities, extended thinking, batches, and the Claude Agent SDK. Use when building applications that call Claude API across different programming languages.2.4kinstalls 470Ios Icon GenThe ios-icon-gen skill Generate iOS app icons as PNG imagesets for Xcode asset catalogs from SF Symbols 5000+ Apple-native or Iconify API 275k+ open source icons from 200+ collections . Use when generating icons, creating icon assets, adding icons to asset catalog, or searching for icons for iOS projects. It covers generating icon assets for an iOS/macOS Xcode project. Key workflows include searching for icons across open source collections. Developers invoke ios-icon-gen when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation.2.4kinstalls 471Make Interfaces Feel BetterTurn flat or generic UI components into polished, tactile interfaces that feel finished and delightful.2.4kinstalls 472Redis PatternsThe redis-patterns skill Redisデータ構造パターン、キャッシング戦略、分散ロック、レート制限、Pub/Sub、本番アプリケーション用コネクション管理。 It covers pub/SubまたはRedis Streams for messaging使用. Key workflows include 本番環境でRedis設定（プール、削除、クラスタリング）. Developers invoke redis-patterns when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation.2.4kinstalls 473Ui To VueThe ui-to-vue skill UIスクリーンショットやデザインエクスポートをVue 3コンポーネントに一括変換する際に使用。Vant、Element Plus、Ant Design Vueに対応。 It covers ユーザーがデザインスクリーンショットまたはデザインエクスポート画像のディレクトリを提供するとき。. Key workflows include ターゲットアプリケーションがVue 3のとき。. Developers invoke ui-to-vue when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, a.2.4kinstalls 474Motion UiThe motion-ui skill 日本語翻訳：このファイルは motion-ui 用の日本語翻訳が必要です It covers description: 日本語翻訳：このファイルは motion-ui 用の日本語翻訳が必要です. Key workflows include このファイルの翻訳は実装中です。英語版は元のスキルファイルを参照してください。. Developers invoke motion-ui when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied fr.2.4kinstalls 475Netmiko Ssh AutomationThe netmiko-ssh-automation skill 日本語翻訳：このファイルは netmiko-ssh-automation 用の日本語翻訳が必要です It covers description: 日本語翻訳：このファイルは netmiko-ssh-automation 用の日本語翻訳が必要です. Key workflows include netmiko-ssh-automation - 日本語翻訳進行中. Developers invoke netmiko-ssh-automation when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configu.2.4kinstalls 476Scholar EvaluationThe scholar-evaluation skill 論文、提案書、文献レビュー、方法論セクション、証拠の質、引用サポート、研究論文フィードバックのための構造化された学術的作業評価。 It covers 研究論文、提案書、論文章、または文献レビューのレビュー。. Key workflows include 主張が引用された証拠によって支持されているかの確認。. Developers invoke scholar-evaluation when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation chec.2.4kinstalls 477Django CeleryThe django-celery skill from everything-claude-code covers background jobs and async processing in Django projects. Use cases include background email send, heavy processing schedules, daily reports, cleanup crons, external API queueing, and multi-step workflow coordination. Setup installs celery and redis, defines shared_task functions, and starts workers with celery -A myapp worker. Patterns show send_email.delay for async queueing and beat_schedule crontab entries for periodic tasks. Error handling checklist covers retry logic, dead letter handling, logging, and Flower monitoring. Published SKILL.md is Japanese language with pip install, shared_task examples, and beat crontab configuration. Agents guide broker selection (Redis/RabbitMQ), worker scaling, and task idempotency for Django services.2.3kinstalls 478Error HandlingComprehensive error handling guide for production applications spanning TypeScript, Python, and Go. Covers typed error hierarchies, Result patterns, API error handlers, React error boundaries, retry mechanisms with exponential backoff, and circuit breaker patterns. Emphasizes fail-fast principles, structured error types over strings, separation of user-facing and developer messages, and proper error propagation. Includes concrete implementations, checklists, and best practices for preventing silent failures, managing external dependencies reliably, and documenting error contracts in APIs.2.3kinstalls 479Cisco Ios PatternsCisco IOS patterns provide a systematic approach to reviewing network device configurations, collecting evidence from routers and switches without incident escalation, and validating planned changes safely. This skill covers show command selection for read-only troubleshooting, configuration hierarchy navigation (global, interface, routing process, line modes), ACL wildcard mask validation, interface hygiene checks, and change window verification workflows. Essential for pre-change reviews, security audits, and incident response on Cisco platforms. Emphasizes capturing baseline state before modifications, verifying administrative access remains open, applying minimal changes during maintenance windows, and validating results before persisting to startup-config.2.3kinstalls 480Skill Stocktakeskill-stocktake is a slash command that systematizes quality audits of Claude skills and commands across global and project-level directories. It supports two modes: Quick Scan (5-10 min) re-evaluates only recently changed skills using a cached results.json; Full Stocktake (20-30 min) performs complete inventory and holistic AI-driven evaluation. Phase 1 scans and catalogs skills with metadata; Phase 2 batches ~20 skills per subagent invocation against a checklist covering overlap, freshness, and actionability; Phase 3 summarizes verdicts (Keep, Improve, Update, Retire, Merge); Phase 4 consolidates changes with user confirmation. Results persist in ~/.claude/skills/skill-stocktake/results.json with resumable in-progress tracking.2.3kinstalls 481Ecc GuideThe ECC Guide skill helps developers navigate Everything Claude Code, a living repository of agents, skills, commands, hooks, and install profiles. It emphasizes real-time filesystem inspection over stale catalogs, enabling users to discover components, understand relationships between skills and commands, and resolve installation paths. Core tasks include onboarding new users, feature discovery, managed install guidance, project-specific configuration via /project-init, and troubleshooting harness integration. The skill prioritizes skills-first workflows while maintaining command compatibility shims and delegates to subagents when appropriate, ensuring current answers grounded in repository state.2.3kinstalls 482Agentic OsThe agentic-os skill helps with Claude Code 上に永続的なマルチエージェントオペレーティングシステムを構築します。カーネルアーキテクチャ、スペシャリストエージェント、スラッシュコマンド、ファイルベースのメモリ、スケジュールされた自動化、外部データベースなしの状態管理をカバーします。. Key workflows include エージェニック OS; 起動タイミング; アーキテクチャ概要; 層の責任. Documented capabilities cover Claude Code 内でマルチエージェントワークフローを構築する; セッション再起動後も維持される永続的な Claude Code 自動化をセットアップする; 繰り返しタスク向けの「パーソナル OS」または「エージェニック OS」を作成する; ユーザーが「エージェニック OS」、「パーソナル OS」、「マルチエージェント」、「エージェントコーディネーター」、「永続エージェント」と言う; コンテキストがセッションをまたいで維持される必要がある長期プロジェクトを構造化する. Agents should invoke it when users ask about agentic os or mention triggers defined in the skill frontmatter. Follow the SKILL.md steps, reference files, and output formats rather than improvising outside documented scope. - Claude Code 内でマルチエージェントワークフローを構築する - セッション再起動後も維持される永続的な Claude Code 自動化をセットアップする - 繰り返しタスク向けの「パーソナル OS」または「エージェニック OS」を作成する - ユーザーが「エージェニック OS」、「パーソナル OS」、「マルチエージェント」、「エージェントコーディネーター」、「永続エージェント」と言う - コンテキストがセッションをまたいで維持される必要がある長期プロジェクトを構造化する ## アーキテクチャ概要エージェニック OS には 4 つの層があります。各層はプロジェクトルートのディレクトリです。 project-root/ ├── CLAUDE.md # カーネル: アイデンティティ、ルーティングルール、エージェントレジストリ ├── agents/ # スペシャリストエージェント定義（Markdown プロンプト） ├── .claude/commands/ # スラッシュコマンド: ユーザー向け CLI ├── scripts/ # デー.2.3kinstalls 483Homelab Network ReadinessThe homelab-network-readiness skill is a read-first planning and review workflow for home or small-lab networks mixing VLANs, Pi-hole or AdGuard DNS, firewall rules, and remote VPN access. It inventories internet edge, gateway, switching, Wi-Fi SSID mapping, addressing conflicts, DHCP/DNS ownership, management paths, and recovery options before recommending changes. Trust-zone tables cover trusted, server, IoT, guest, management, and VPN segments with default-deny policies between zones. DNS filtering readiness stages Pi-hole or similar resolvers with reserved addresses, fallback paths, and per-VLAN tests before broad DHCP rollout. Remote access guidance compares split tunnel, full tunnel, and overlay VPN modes with port-forwarding guardrails. The skill refuses copy-paste router commands without confirmed platform, rollback, console access, and maintenance windows. Anti-patterns flag moving admin workstations off reachable management networks or pointing every DHCP scope at a new resolver without fallback.2.3kinstalls 484Network Config ValidationThe network-config-validation skill from everything-claude-code helps agents review network configuration artifacts such as router, switch, firewall, and host network settings before deployment or during incident response. Although the published SKILL.md is a Japanese translation stub pointing to the upstream ECC skill, the intended workflow validates configuration syntax, checks for hazardous patterns like open ACLs or missing encryption, compares files against known-good baselines, and summarizes diffs operators should fix. Agents use it when users paste configs, reference network-config-validation, or ask to audit interface, routing, VLAN, DNS, or firewall stanzas prior to apply. Output should flag blocking errors separately from warnings and cite exact stanza lines when possible. Pair with network-bgp-diagnostics when BGP session issues appear after config changes.2.3kinstalls 485Flox EnvironmentsThe flox-environments skill guides agents through Nix-backed Flox setup for reproducible project-scoped environments on macOS and Linux. It covers flox init, search, install, activate, and manifest.toml sections for packages, vars, hooks, profiles, services, and platform systems. Language recipes include Python with uv venvs, Node.js, Rust, Go, and C/C++ with pkg-group and priority patterns. Hooks must stay idempotent using FLOX_ENV_CACHE guards; profiles hold interactive shell helpers. The skill explains git-native sharing via .flox commits, FloxHub push and include composition, and agent-friendly flox install without sudo. Anti-patterns cover absolute paths, exit in hooks, secrets in manifests, and slow non-idempotent installs. Full-stack examples wire PostgreSQL and Redis services with on-activate database init. Use when users need declarative toolchains, local services, or fixes for works-on-my-machine drift across teammates and CI.2.3kinstalls 486Motion AdvancedThe motion-advanced skill from everything-claude-code covers advanced Motion library techniques for web interfaces beyond basic fades and slides. The catalog SKILL.md is a Japanese translation stub referencing upstream ECC, but the skill intent guides agents through layout animations, shared element transitions, staggered children, scroll-linked motion, drag and gesture handlers, spring tuning, reduced-motion accessibility, and avoiding layout thrash on large lists. Use when polishing interactive dashboards, marketing pages, or component libraries that need choreographed enter-exit sequences without jank. Agents should profile animation cost, respect prefers-reduced-motion, and keep transforms GPU-friendly. Pair with frontend-design skills when motion must reinforce a committed visual identity rather than generic transitions.2.3kinstalls 487Network Bgp DiagnosticsThe network-bgp-diagnostics skill from everything-claude-code guides agents through systematic Border Gateway Protocol troubleshooting when peers flap, sessions stay idle, or prefixes stop propagating. The catalog SKILL.md is currently a Japanese translation stub referencing the upstream ECC source, but the skill intent is to interpret show bgp summary, neighbor, and advertised-routes output, verify AS path and prefix filters, check hold timers and authentication mismatches, and correlate recent configuration pushes with session loss. Agents collect neighbor state, last error codes, and policy maps, then recommend corrective steps such as fixing MD5 keys, TTL security, max-prefix limits, or export route maps. Use alongside network-config-validation when a config change preceded the outage. Deliver concise root-cause hypotheses ranked by likelihood with verification commands.2.3kinstalls 488Motion PatternsThe motion-patterns skill from everything-claude-code teaches reusable Framer Motion (Motion) animation patterns for React applications. The published catalog SKILL.md is a Japanese translation stub referencing the upstream ECC motion-patterns source, but the intended skill covers pattern libraries for enter/exit transitions, staggered lists, layoutId shared element transitions, gesture-driven interactions, and spring-based motion tuned for GPU-friendly transform and opacity animations. Agents guide users away from layout-thrashing width/height animations toward x, y, scale, rotate, and opacity with will-change hints, useMemo variants, and AnimatePresence mount cycles. Use when users reference motion-patterns, need animation recipes, or want Motion/React motion UX patterns beyond basic fades.2.3kinstalls 489Tinystruct PatternsThe tinystruct-patterns skill helps with Expert guidance for developing with the tinystruct Java framework. Use when working on the tinystruct codebase or any project built on tinystruct - including creating Application classes, @Action-mapped routes, unit tests, ActionRegistry, HTTP/CLI dual-mode handling, the built-in HTTP server, the event system, JSON with Builder/Builders, database persistence with AbstractData, POJO generation, Ser. Key workflows include tinystruct Development Patterns; Core Principle; When to Activate; When to Use. Documented capabilities cover Creating new `Application` modules by extending `AbstractApplication`.; Defining routes and command-line actions using `@Action`.; Handling per-request state via `Context`.; Performing JSON serialization using the native `Builder` and `Builders` components.; Working with database persistence via `AbstractData` POJOs.. Agents should invoke it when users ask about tinystruct patterns or mention triggers defined in the skill frontmatter. Follow the SKILL.md steps, reference files, and output formats rather than improvising outside documented scope.2.3kinstalls 490Plan OrchestrateThe plan-orchestrate skill from everything-claude-code coordinates long-horizon agent work by decomposing goals into ordered sub-plans, assigning specialized subagents, and enforcing review checkpoints before merges or deploys. The catalog SKILL.md is currently a Japanese translation stub referencing the upstream ECC source, but the intended workflow breaks ambiguous requests into verifiable milestones, tracks dependencies between exploration, implementation, and verification phases, and surfaces blockers early instead of monolithic execution. Agents use it when users need structured plan execution across repositories, tests, and documentation updates with explicit handoff criteria between steps. Output should include a living plan graph, status per subtask, and rollback or retry guidance when a step fails validation. Pair with review and testing skills when plans touch production paths.2.3kinstalls 491Motion FoundationsThe motion-foundations skill from everything-claude-code introduces Framer Motion (Motion) basics for React UIs. The catalog SKILL.md is a Japanese translation stub pointing to the upstream ECC motion-foundations source, but the intended curriculum covers installing Motion, motion component syntax, animate and initial props, transition timing, AnimatePresence mount/unmount cycles, and foundational transform versus layout property guidance before advanced motion-patterns recipes. Agents teach GPU-friendly opacity and transform animations, when to reach for springs, and how to structure simple enter/exit effects without layout thrash. Pair with motion-patterns for reusable animation recipes after foundations. Use when users need Motion getting started, animation basics, or ECC motion-foundations onboarding.2.3kinstalls 492Quarkus PatternsThe quarkus-patterns ECC skill documents cloud-native Quarkus architecture with JAX-RS or RESTEasy Reactive resources, service layers using ApplicationScoped CDI, Hibernate Panache repositories, Apache Camel RabbitMQ publishers, validation, exception mapping, pagination, and profile-specific YAML config. Examples show OrderProcessingService with RequiredArgsConstructor injection, transactional boundaries, async fulfillment publishers, and custom LogContext MDC patterns. Activation covers event-driven flows, CompletableFuture async work, GraalVM native compilation, and environment profiles for dev staging prod. Catalog SKILL content is Spanish-language with upstream ECC origin. Use when building Quarkus APIs, structuring resource-service-repository layers, or integrating Camel messaging in JVM microservices.2.3kinstalls 493Angular DeveloperThe angular-developer skill from everything-claude-code provides architectural guidance and code generation for Angular projects. Activation covers scaffolding apps and libraries, components, services, directives, pipes, guards, signal-based reactivity with linkedSignal and resource, signal and reactive forms, dependency injection, lazy routing, ARIA accessibility, animations, styling with component styles or Tailwind, and unit or E2E tests. It requires analyzing the project Angular version before advice and running ng build after code generation to catch errors. ng new execution rules choose npx with explicit versions, local ng when installed, or npx latest as fallback. Reference files split components, inputs, outputs, signals, forms, routing, SSR, testing, and CLI tooling. Prefer signal forms on supported versions for new work while matching legacy form strategies on older apps.2.3kinstalls 494Homelab Wireguard VpnThe homelab-wireguard-vpn skill configures WireGuard for fast modern remote access to home networks on Raspberry Pi, Linux hosts, pfSense, or routers. Explains encrypted UDP tunnel on port 51820, per-device keypairs, server peer public keys, and client endpoint IP port configuration without central certificate authority. Linux server setup covers apt install wireguard, umask 077 key generation, wg0.conf interface addressing, iptables forwarding and NAT rules, and systemctl enable wg-quick@wg0. Client config templates document AllowedIPs for split tunnel home subnets versus full tunnel routing. Troubleshooting addresses handshake failures, firewall UDP 51820 blocks, and key permission issues. Automating peer generation supports multiple phones and laptops. Invoke for homelab VPN, remote NAS access, split tunneling guidance, or WireGuard connection debugging. WireGuard UDP tunnel on port 51820 with per-device keypairs. Linux server setup: wg genkey, wg0.conf, iptables NAT forwarding.2.3kinstalls 495Quarkus SecurityThe quarkus-security skill documents security best practices for Quarkus applications covering authentication, authorization, input validation, and secrets management. It guides JWT and OIDC setup with @Authenticated resources, JsonWebToken injection, and application.properties issuer configuration. Authorization patterns use @RolesAllowed and SecurityIdentity checks. Bean Validation and custom validators protect user input; CORS and security headers harden HTTP surfaces. Secrets integrate via Vault or environment configuration sources with guidance on rate limiting and brute-force protection. Dependency scanning addresses CVE exposure in MicroProfile JWT and SmallRye JWT stacks. Trigger when adding auth, configuring OIDC realms, validating input, managing secrets, or scanning Quarkus dependencies for vulnerabilities during security reviews. JWT and OIDC authentication with @Authenticated and JsonWebToken. RBAC via @RolesAllowed and SecurityIdentity role checks. Bean Validation and custom input validators for user data. Secrets via Vault or env config sources with rate limiting guidance.2.3kinstalls 496Quarkus TddThe quarkus-tdd skill guides test-driven development for Quarkus 3.x LTS services targeting eighty percent plus unit and integration coverage with JUnit 5, Mockito, REST Assured, Apache Camel tests, and JaCoCo reporting. Workflow is red-green-refactor: write failing tests first, implement minimal code, refactor with green suite. Covers REST endpoints, data access, security rules, reactive streams, Camel routes, RabbitMQ event handlers, conditional flows, CompletableFuture async ops, and LogContext propagation. Unit tests use @Nested organization and MockitoExtension; integration tests leverage @QuarkusTest and REST Assured. Invoke when adding Quarkus features, fixing bugs, refactoring event-driven services, or enforcing coverage gates on Camel and messaging code paths. Red-green-refactor TDD workflow for Quarkus 3.x LTS. JUnit 5, Mockito, REST Assured, and Camel route testing. JaCoCo coverage target 80 percent plus unit and integration. @Nested unit test organization with MockitoExtension.2.3kinstalls 497GgetThe gget skill from everything-claude-code guides agents through lightweight genomics lookups via the gget CLI or Python package before heavier Biopython or Nextflow pipelines. Modules cover search, info, seq, ref, blast, blat, muscle, diamond, alphafold, pdb, enrich, and related database queries with small-query-first discipline. Installation uses clean venv or uv environments with documented upgrade checks because upstream databases change. Agents record module name, version, arguments, and output filenames for reproducible evidence logs. Scope excludes regulated clinical interpretation and high-throughput production pipelines needing fine-grained index control. Use for Ensembl ID lookup, quick BLAST, reference genome links, and exploratory bioinformatics validation.2.3kinstalls 498Mle WorkflowThe mle-workflow ECC skill turns model work into production ML with explicit data contracts, repeatable training, measurable promotion gates, deployable artifacts, and operational monitoring. Activation covers ranking, classifiers, recommenders, forecasting, embeddings, and LLM workflows without forcing one architecture. Ten task simulations from product framing through refresh define contracts, baselines, features, thresholds, error analysis, artifact packaging, serving, canary rollout, and post-launch drift response. It reuses SWE surfaces like TDD, eval-harness, deployment-patterns, and security-review rather than isolated notebook culture. Scope calibration warns against assuming labels, feature stores, or A/B tests exist. Use when building, reviewing, or hardening ML beyond one-off experiments.2.3kinstalls 499Uspto DatabaseThe uspto-database ECC skill workflows official US patent and trademark records through USPTO Open Data Portal, Patent File Wrapper, PatentSearch API, TSDR, and assignment search surfaces. Agents prefer official sources over secondary indexes, store API keys in environment variables never in commits, and log endpoints, query bodies, dates, and result counts for reproducibility. PatentSearch JSON queries filter by dates, assignees, and fields with pagination; trademark status uses TSDR; ownership transfers use assignment search. Secondary tools like Google Patents are cross-check only. Skill is for data collection not legal advice. Use for prior art search, application status, portfolio landscape, or trademark record verification.2.3kinstalls 500Quarkus VerificationThe quarkus-verification skill defines a verification loop for Quarkus projects before pull requests, major refactors, or staging and production deploys. Phases include Maven or Gradle clean verify builds, static analysis, tests with 80%+ coverage thresholds, security scans, native image compatibility checks, and pre-merge diff review. Activation covers dependency updates, pre-deploy gates, and validating that test coverage and native builds pass. Commands reference mvn clean verify patterns and native compilation steps typical in Quarkus microservices. Agents run the full pipeline rather than skipping security or native phases. Use when users prepare Quarkus PRs, validate coverage gates, or verify native image builds before release. Full pipeline: build, lint, tests, security scan, native compile, diff review 80%+ coverage threshold validation before merge or deploy Activates before PRs, major refactors, and staging or production releases Maven and Gradle verify commands with native image compatibility checks Prevents skipping security scans or native build validation in Quarkus services quarkus-verification runs the full Quarkus build-test-security-native verification loop befor.2.3kinstalls 501Fsharp TestingThe fsharp-testing skill from everything-claude-code documents comprehensive F# testing patterns using xUnit, FsUnit.xUnit, Unquote quotations, and FsCheck.xUnit property-based tests. It activates when writing new F# tests, reviewing coverage, setting up test infrastructure, or debugging flaky suites. The stack table maps each library to purpose: xUnit as framework, FsUnit for idiomatic assertions, Unquote for quotation-based failure messages, and FsCheck for generative properties. Guidance covers test organization, async testing, mocking boundaries, and modern .NET practices for functional code. Use when users test F# modules, add property tests, or improve assertion clarity in .NET solutions. xUnit, FsUnit, Unquote, and FsCheck.xUnit as the core F# test stack Property-based testing patterns integrated with xUnit runners Activates for new tests, coverage review, infrastructure setup, and flaky test fixes F#-friendly assertion syntax and quotation-based failure messages Modern .NET async and integration testing practices for functional code fsharp-testing provides F# unit and property-based testing patterns with xUnit, FsUnit, Unquote, and FsCheck Organized F# test suites with idi.2.3kinstalls 502Cost TrackingThe cost-tracking skill analyzes Claude Code cost and usage history from a local SQLite database at ~/.claude-cost-tracker/usage.db written by cost tracking hooks or plugins. Japanese SKILL.md documents activation when users ask how much they spent, session cost, token usage, budget limits, overages, or breakdowns by project, tool, session, model, or date including today vs yesterday trends and CSV export of recent records. Agents verify prerequisites exist before querying. Rescued from community PR #1304. Use when users track Claude Code spend locally rather than cloud billing dashboards alone. Queries local SQLite usage.db for Claude Code token and cost history Breakdowns by project, tool, session, model, and date ranges Budget, spend limit, and overage reporting when data exists Today vs yesterday comparisons and recent usage CSV export Requires cost tracking hook writing to ~/.claude-cost-tracker/usage.db cost-tracking reports Claude Code usage and spend from the local cost-tracker SQLite database Cost and token report by session, project, tool, model, or date from local DB User asks session cost, token usage, budget, or Claude Code spend2.2kinstalls 503Skill ScoutThe skill-scout everything-claude-code skill searches local installs, marketplaces, GitHub, and the web for existing skills before authors build duplicates. It supports create, build, fork, and search workflows with Japanese and English trigger guidance. Agents present closest matches, license notes, and fork-versus-new recommendations. Use when users plan new skills or ask whether a capability already exists. Searches local, marketplace, GitHub, and web skills. Recommends fork versus new skill creation. Supports create, build, fork, and search flows. Surfaces closest matches and license notes. Avoids duplicate skill authoring effort. Search local, marketplace, GitHub, and web for existing skills before creating new ones.2.2kinstalls 504Homelab Vlan SegmentationThe homelab-vlan-segmentation skill splits flat home networks into isolated VLANs preventing IoT guests and trusted PCs mutual access. Documents VLAN templates trusted IoT servers guest management subnets example 192.168 ranges trunk access ports SSID-to-VLAN mapping inter-VLAN firewall on pfSense OPNsense UniFi. Apply in maintenance windows verifying connectivity after each step without removing existing protections. Use first VLAN setup isolating smart home guest Wi-Fi or troubleshooting inter-VLAN routing for homelab security upgrades. VLANs trusted IoT servers guest management. Subnet templates and gateways. Trunk access SSID VLAN mapping. Firewall blocks IoT guest from trusted. pfSense OPNsense UniFi support. Homelab VLAN segmentation guide. User asks home VLAN IoT isolation.2.2kinstalls 505Homelab Pihole DnsThe homelab-pihole-dns skill configures Pi-hole network-wide DNS ad blocking on Raspberry Pi or Linux host. Documents normal versus Pi-hole DNS flow blocking tracker domains with null responses forwarding allowed queries to upstream Cloudflare Google or DoH resolvers. Covers installation blocklist management local DNS records like nas.home.lan troubleshooting devices losing internet after Pi-hole and running alongside DHCP. Every device gets blocking without browser extensions. Use installing Pi-hole configuring home DNS adding blocklists or creating local homelab name resolution. Network-wide DNS ad blocking via Pi-hole. Blocklists return null for ads trackers. Upstream DoH or public resolvers. Local DNS records for homelab hosts. Troubleshoot post-install connectivity. Pi-hole homelab DNS setup. User asks Pi-hole DNS homelab.2.2kinstalls 506Prisma Patternsprisma-patterns is an agent skill from affaan-m/everything-claude-code that prisma orm patterns for typescript backends — schema design, query optimization, transactions, pagination, and critical traps like updatemany returning count not records, $transaction timeouts, migrat. # Prisma Patterns Production patterns and non-obvious traps for Prisma ORM in TypeScript backends. Tested against Prisma 5.x and 6.x. Some behaviors differ from Prisma 4. Check the Prisma version before applying version-specific patterns: ```bash npx prisma --version ``` Prisma 5 introduced `relationJoins`, which can load relations via JOIN rat Developers invoke prisma-patterns during operate/infra work for cloud & infrastructure tasks. The skill documents triggers, prerequisites, and step-by-step workflows grounded in SKILL.md. Compatible with Claude Code, Cursor, and Codex agent runtimes that load marketplace skills. Review the Security Audits panel on this listing before installing in production environments.2.1kinstalls 507Blender Motion State InspectionThe blender-motion-state-inspection skill is designed for use this skill when inspecting Blender characters, rigs, poses, animation retargeting, ground contact, facing direction, or model-vs-motion alignment where screenshots alone. Blender Motion State Inspection When to Use A Blender character looks twisted, mirrored, flattened, offset, or foot-sliding in an animation. A user asks whether an imported avatar, armature, or retargeted motion matches an expected pose. Invoke when the user asks about blender motion state inspection or related SKILL.md workflows.2kinstalls 508Recsys Pipeline ArchitectThe recsys-pipeline-architect skill is designed for design composable recommendation, ranking, and feed pipelines using the six-stage Source→Hydrator→Filter→Scorer→Selector→SideEffect framework popularized by xAI's open-sourced. recsys-pipeline-architect A spec-and-scaffold skill for building composable recommendation, ranking, and feed pipelines. It encodes the six-stage pattern — Source → Hydrator → Filter → Scorer → Selector → SideEffect — popularized by xAI's open-sourced For You algorithm (Apache 2.0). Invoke when the user asks about recsys pipeline architect or related SKILL.md workflows.1.9kinstalls 509UncloudThe uncloud skill use when managing an Uncloud cluster - deploying services, configuring Caddy ingress, adding static proxy routes for non-cluster devices, publishing ports, scaling, inspecting logs, or managing machines and volumes with the `uc` CLI. # Uncloud Cluster Management Reference for the `uc` CLI - a decentralised self-hosting platform using Docker containers, WireGuard mesh networking, and Caddy reverse proxy. ## When to Activate Use this skill when working with Uncloud clusters, especially when: - Bootstrapping or joining machines with `uc machine` - Deploying services from Compose files with `uc deploy` - Publishing HTTP, HTTPS, TCP, or UDP ports through Uncloud - Configuring Caddy ingress with `x-caddy`, `x-ports`, or `--caddyfile` - Routing external LAN devices through the cluster proxy - Inspecting logs, service state, volumes, DNS, or machine placement ## How It Works Uncloud runs Docker services across peer machines connected by a WireGuard mesh. Each machine is an equal cluster member; services communicate on the overlay network and Caddy runs globally to terminate public HTTP/HTTPS traffic.1.8kinstalls 510Project Guidelines ExampleAn example skill demonstrating how to encode project-specific guidelines for AI coding agents inside a SKILL.md file. It models conventions, directory layout, workflow expectations, and guardrails teams can copy when authoring their own project-guidelines skill. Use it as a reference template rather than a runtime automation tool.1.8kinstalls 511Frontend DesignBuilds frontend pages and components with deliberate visual design rather than generic default styling. It covers layout, typography, color, and motion for web interfaces. A solo builder reaches for it when a SaaS landing page or app screen needs to look polished quickly.1.8kinstalls 512Project Guidelines Exampleプロジェクトガイドラインスキル例これはプロジェクト固有のスキルの例です自分のプロジェクトのテンプレートとして使用してください実際の本番アプリケーションに基づいています Zenith https zenith chat AI駆動の顧客発見プラットフォーム使用するタイミングこのスキルが設計された特定のプロジェクトで作業する際に参照してくださいプロジェクトスキルには以下が含まれますアーキテクチャの概要ファイル構造コードパターンテスト要件デプロイメントワークフローアーキテクチャの概要技術スタックフロントエンド Next js 15 App Router TypeScript React バックエンド FastAPI Python Pydanticモデルデータベース Supabase PostgreSQL AI Claudeツール呼び出しと構造化出力付きAPI デプロイメント Google Cloud Run テスト Playwright E2E pytest バックエンド React Testing Library サービス Frontend Next js 15 TypeScript TailwindCSS Deployed Vercel Cloud Run Backend FastAPI Python 3 11 Pydantic Deployed Cloud Run Supabase Claude Redis Database API Cache ファイル構造 project frontend src app Next js app routerページ api APIルート auth 認証保護されたルート workspace メインアプリワークスペース components Reactコンポーネント ui ベースUIコンポーネント forms フォームコンポーネント layouts レイアウトコンポーネント hooks カスタムReactフック lib ユーティリティ types TypeScript定義 config 設定 backend routers FastAPIルートハンドラ models py Pydanticモデル main py FastAPIアプリエントリ auth_system py 認証 database py データベース操作 services ビジネスロジック tests pytestテスト deploy デプロイメント設定 docs ドキュメント scripts ユーティリティスクリプトコードパターン APIレスポンス形式 FastAPI python from pydantic import BaseModel from typing import Generic TypeVar Optional.1.7kinstalls 513React Testingreact-testing is an Everything Claude Code skill for behavior-focused React tests using React Testing Library with Vitest or Jest, MSW for network mocking, and axe for accessibility assertions. The skill explains when to write component tests versus Playwright or Cypress end-to-end runs, how to test custom hooks, and how to migrate away from legacy Enzyme patterns. Developers reach for react-testing while adding coverage to untested components, fixing flaky UI tests, or asserting accessible behavior during feature work. It emphasizes user-visible behavior over implementation details and documents practical mocking boundaries for modern React codebases.1.6kinstalls 514React Performancereact-performance is a frontend agent skill adapted from Vercel Engineering's react-best-practices (MIT, v1.0.0) in the vercel-labs/agent-skills repository. It organizes more than 70 rules across eight priority categories: waterfalls, bundle size, server-side rendering, client fetching, re-renders, rendering, JavaScript micro-performance, and advanced patterns for React 18/19 and Next.js. Agents invoke it while writing new components, reviewing pull requests, or refactoring slow pages so fixes follow a severity-ordered checklist instead of ad-hoc tweaks. Developers reach for react-performance when Core Web Vitals regress, bundle analyzers show bloat, or client-side waterfalls appear in React Server Component boundaries.1.6kinstalls 515React Patternsreact-patterns is an agent skill from affaan-m/everything-claude-code that encodes idiomatic React 18/19 guidance for hooks discipline, server and client component boundaries, Suspense with error boundaries, form actions, data fetching, state management decision trees, and accessibility-first composition. Developers reach for react-patterns when writing or modifying React function components, reviewing JSX/TSX files, migrating class or useEffect-heavy code, or choosing state and composition strategies. The skill activates during component design and code review rather than project setup or release automation.1.6kinstalls 516Frontend A11yfrontend-a11y is a community-origin everything-claude-code skill with practical accessibility patterns for React and Next.js applications. The skill addresses issues commonly flagged in code review: missing form labels on input, select, and textarea elements, incorrect ARIA usage, non-semantic interactive elements, and broken keyboard navigation on modals and dropdowns. Developers reach for frontend-a11y when building or reviewing interactive components that must support screen readers and keyboard-only users without a separate audit pass. Activation triggers include new form components, modals, disclosure widgets, and any interactive UI where WCAG-oriented semantics and focus traps matter.1.5kinstalls 517Benchmark Optimization Loopbenchmark-optimization-loop is a performance engineering skill from Everything Claude Code for developers who hear requests like "make it 20x faster" or "try dozens of recursive optimizations" and need a disciplined measurement loop instead of random tweaks. The skill requires a defined operation, a correctness gate that must stay green, and an explicit metric such as wall time, p95 latency, rows per second, cost per run, or memory before any change lands. Agents use Read, Write, Edit, Bash, Grep, and Glob to implement variants, run repeated benchmarks, and select the best measured implementation. Developers reach for this skill when optimization must be evidence-backed rather than intuition-driven.1.5kinstalls 518Marketing Campaignmarketing-campaign is an end-to-end launch planning skill from the Everything Claude Code collection that expands a single product brief into audience research, positioning, campaign angles, landing page copy, email sequences, social posts, ad copy, short-form video scripts, and a content calendar. It acts as the orchestration layer so messaging stays consistent across channels instead of generating one-off posts. Developers reach for marketing-campaign when planning a product or feature launch, defining positioning before copy, or building a full content suite from one brief. The skill emphasizes campaigns that convert, not just assets that ship, and originates from the ECC metadata bundle.1.5kinstalls 519Latency Critical Systemslatency-critical-systems is an ECC-origin skill from affaan-m/everything-claude-code for engineering latency-sensitive systems such as realtime dashboards, market data feeds, streaming agents, execution gateways, queues, and caches. The readme instructs agents to split metrics instead of collapsing on vague speed goals—tracking p50, p95, and p99 latency alongside throughput and freshness. It covers hot-path profiling, queue depth, cache behavior, and streaming freshness for HFT-like infrastructure while explicitly excluding live trading or financial advice. Developers invoke it when p95 regressions, stale dashboards, or agent response delays threaten SLA commitments. Tools listed include Read, Write, Edit, Bash, Grep, and Glob for codebase and benchmark investigation.1.5kinstalls 520Parallel Execution Optimizerparallel-execution-optimizer is a Claude Code skill for speeding up agent work by running independent tasks concurrently without losing correctness. Before acting, it converts urgency into a dependency graph, splits work into lanes, labels each lane parallel, serial, or gated, batches independent reads and checks, and isolates writes by file, worktree, branch, service, or dataset. A compact lane matrix records whether each lane can run in parallel, its write surface, risk, and verification method. Developers use it for large repo scans, simultaneous backend and frontend patches in separate worktrees, or gated deploy readbacks. The SKILL.md body is written in Chinese.1.5kinstalls 521Social Publishersocial-publisher connects Claude Code to SocialClaw for agent-driven social media publishing across 13 platforms through a single workspace API key. A developer builds a schedule.json of posts, validates it, then applies and monitors delivery run status with the socialclaw CLI. It handles X, LinkedIn, Instagram, Facebook Pages, TikTok, YouTube, Reddit, WordPress, Discord, Telegram, and Pinterest. It is used when a launch announcement or campaign needs coordinated cross-platform distribution.1.5kinstalls 522Continuous Learning V2continuous-learning-v2 is an instinct-based learning system from affaan-m/ecc (Everything Claude Code) at version 2.1.0. It watches Claude Code sessions via hooks, records small learned behaviors as atomic instincts with confidence scores, and promotes mature instincts into skills, slash commands, and subagents. Version 2.1 adds project-scoped instincts so React conventions stay in React repos and Python patterns stay in Python repos, preventing cross-project contamination. Developers reach for continuous-learning-v2 when repeated session patterns should compound into reusable agent knowledge instead of being re-explained each chat. The architecture treats instincts as evolvable units rather than one-off prompt snippets.1.5kinstalls 523Data Throughput Acceleratordata-throughput-accelerator is a Claude Code skill for making large data ingestion, backfill, export, ETL, and table synchronization jobs much faster while preserving correctness. It separates source extraction, network transfer, warehouse load, transform speed, serving-table freshness, and live-tail growth before applying optimizations. A data engineer uses it when a pipeline is correct but too slow, especially when new data arrives faster than the catch-up window. It matters because it prioritizes provably correct faster landing over raw speed alone.1.5kinstalls 524Recursive Decision Ledgerrecursive-decision-ledger is a Claude Code skill for developers forcing deeper computation through repeated rollouts or Prime Gauss style recursive prompting. It preserves repeated trials, prior memory, fresh information, and explicit marks while removing false certainty, recording each rollout under a ledger contract (rollout id, prior winner, fresh info, search space size, trial counts, decision and coherence marks). It prefers append-only JSONL ledgers with Markdown summaries. Developers use it when building agents that compare ensemble outputs, explore local optima, or run stochastic optimization and need an auditable decision log.1.5kinstalls 525Prediction Market Risk Reviewprediction-market-risk-review is a Claude Code skill that reviews prediction-market, oracle, and trading-agent workflows for compliance, safety, data quality, privacy, and execution risk before they handle venue authentication, user portfolio data, API keys, or trade planning. It walks structured review gates covering the advice boundary (strip buy/sell/hold/size recommendations), a venue and regulatory boundary, data quality, security (no private keys, read-only scopes, keys out of logs), and privacy. Developers use it as a compliance checklist before shipping agents, dashboards, or APIs that integrate prediction-market venues.1.4kinstalls 526Prediction Market Oracle Researchprediction-market-oracle-research is a Claude Code skill for evaluating prediction markets as data sources or oracle-like signals for products, agents, and decision intelligence. It runs source-grounded analysis of market-implied probabilities, records them with timestamps and source links, and scores signal quality on liquidity, spread, market age, incentive concentration, and resolution authority. It explicitly gives no investment advice and never treats market prices as objective truth. Developers use it when designing agents, dashboards, or decision tools that might consume Polymarket-style probability feeds.1.4kinstalls 527Agent Harness Constructionagent-harness-construction is an ECC skill for developers building AI agent harnesses that constrain output quality through four pillars: action-space design, observation quality, recovery quality, and context-budget quality. The skill prescribes stable explicit tool names, schema-first narrowed inputs, deterministic output shapes, and avoidance of catch-all tools except when isolation is impossible. Granularity rules separate micro-tools for high-risk deploy and migration operations, medium tools for edit-read-search loops, and macro-tools only when round-trip overhead dominates. Every tool response should expose status, summary, next_actions, and artifacts, while error paths follow a structured recovery contract. Developers reach for agent-harness-construction when agent completion rates stall despite capable models, usually because tool design—not model choice—is the bottleneck.1.4kinstalls 528Autonomous Agent Harnessautonomous-agent-harness is an Everything Claude Code skill for building self-directing agent systems without external frameworks like Hermes or AutoGPT. The skill composes Claude Code native crons, dispatch, MCP servers, and persistent memory to queue tasks, run scheduled operations, and maintain state across sessions while enforcing explicit user consent and safety boundaries before scheduling, remote dispatch, credential use, or external posting. Developers reach for autonomous-agent-harness when they need continuous automated operations, scheduled background tasks, or self-directed agent loops scoped to an approved workspace. The skill prefers dry-run plans and local queue files before enabling credential access or modifying third-party resources.1.4kinstalls 529Team Agent Orchestrationteam-agent-orchestration is a Claude Code skill for running team-based orchestration across agent squads instead of a single assistant. It structures work into work items with explicit ownership, an agent Kanban with defined columns and exit criteria, branch and worktree isolation, control-pane visibility, and merge gates. Each card follows a JSON schema with owner, state, branch, acceptance criteria, and merge gate. Developers use it when a task spans multiple agents, harnesses, branches, or worktrees and existing fan-out produces output but not mergeable product.1.2kinstalls 530Dynamic Workflow Modedynamic-workflow-mode is a Claude Code skill for designing task-local agent harnesses, eval gates, and reusable skill extraction. It helps a coding agent generate or adapt a custom loop, evaluator, crawler, or watcher for one task instead of only following a static command flow. A developer uses it when a task needs a custom harness, when multiple agents need the same repeatable process, or when work needs durable handoff artifacts and operator approval before merge. It matters because it turns ad-hoc agent scripting into a disciplined system with eval and merge gates.1.2kinstalls 531Kubernetes Patternskubernetes-patterns is a Claude Code skill providing production-grade Kubernetes patterns for deploying, managing, and debugging workloads. It supplies copy-pasteable YAML for Deployments, Services and Ingress, ConfigMaps and Secrets, RBAC, HPA and PodDisruptionBudgets, and Jobs and CronJobs, plus a decision table for liveness, readiness, and startup probes and a kubectl debugging cheatsheet. A developer uses it when writing, reviewing, or debugging Kubernetes manifests and diagnosing pod failures like CrashLoopBackOff or OOMKilled.1.1kinstalls 532Codehealth Mcpcodehealth-mcp is a Claude Code skill that drives the CodeScene Code Health MCP server to give structural, design-level maintainability feedback for AI-assisted coding. It reviews a file before edits, checks the score delta after changes, and gates commits and pull requests against regressions. A developer uses it when reviewing code quality, refactoring, or checking whether an AI diff degraded a file. It complements lint and style skills with design-level health scores rather than replacing tests.1.1kinstalls 533Intent Driven Developmentintent-driven-development is a Claude Code skill that converts ambiguous or high-impact product and engineering requests into scoped, verifiable acceptance criteria before or alongside implementation. It inspects the repo, docs, and tests first, asks only the questions it cannot infer, and writes observable AC-NNN criteria each with a starting condition, trigger, expected outcome, prohibited side effect, verification method, and priority. A developer uses it to de-risk security, data, migration, or integration changes or to prepare a testable handoff for another agent.1.1kinstalls 534Orch Pipelineorch-pipeline is the shared orchestration engine that the orch-* operation skills (orch-add-feature, orch-fix-defect, orch-refine-code, orch-build-mvp) delegate to. It defines a six-phase Research-Plan-TDD-Review-Commit pipeline, a size classifier that scales process to the change's blast radius, an agent/command map, and two human approval gates. Developers use it indirectly when running an orch-* skill, or directly when adding a new operation to the family. It matters because it gives one consistent, gated umbrella over multiple coding operations.1.1kinstalls 535Inherit Legacy Styleinherit-legacy-style is a Claude Code skill that stops an AI agent from imposing its own idioms on a hand-written legacy project. It scans the codebase across four meta-architecture dimensions (file anatomy, state and control flow, infrastructure, error handling), resolves strong-signal conflicts with the user one question at a time, and writes an enforceable .ai-style-rules.md with Golden Files, naming rules, and DONTs. A developer runs it when onboarding an agent onto an existing project to prevent style drift, and it can install a soft or hard enforcement hook.1.1kinstalls 536Orch Build Mvporch-build-mvp is a Claude Code skill that orchestrates bootstrapping a working MVP from a design or spec document. It ingests an SDD or PRD, orders the feature list into thin vertical slices, scaffolds the first end-to-end slice, then TDD-implements, reviews, and gated-commits each slice, reusing the existing GAN harness to run a generator-evaluator loop against the spec. A developer uses it to turn a spec document into a running starting point with slice-plan and pre-commit gates.1.1kinstalls 537Orch Fix Defectorch-fix-defect is a Claude Code skill that orchestrates fixing a bug by reproducing it as a failing regression test, fixing to green, reviewing, and gated-committing. It is a thin wrapper over the shared orch-pipeline engine that delegates each phase to the matching ECC agent, scoping unclear root causes with code-explorer and escalating build breaks to build-error-resolver. A developer uses it when existing behavior is broken (wrong output, an error, a crash, or a regression), distinct from changing or adding a feature.1.1kinstalls 538Orch Add Featureorch-add-feature is a Claude Code skill that orchestrates building a brand-new feature end to end by delegating each phase to the matching ECC agent. It is a thin wrapper over the shared orch-pipeline engine, running research, plan, TDD implementation, review, and a gated commit, and it classifies size to right-size trivial features. A developer uses it to add net-new behavior that does not exist yet, distinct from fixing a defect or changing existing behavior, with plan-approval and pre-commit gates.1.1kinstalls 539Orch Refine Codeorch-refine-code orchestrates a behavior-preserving refactor. It confirms the existing tests are green, restructures the code without changing behavior, keeps the tests green through small steps, runs a code review, then commits behind a gate. It is a thin wrapper over the shared orch-pipeline engine and delegates dead-code and duplication sweeps to the refactor-cleaner agent. Developers use it when structure should improve but behavior must not change.1.1kinstalls 540Orch Change Featureorch-change-feature is a Claude Code skill that orchestrates altering an existing, working feature to new desired behavior. It updates the existing tests to express the new spec first, then changes the implementation until they pass, reviews, and gated-commits, running through the shared orch-pipeline engine. A developer uses it when behavior is not broken but should be different, distinct from adding a new feature or fixing a defect, keeping the plan light for small tweaks.1.1kinstalls 541Config Gcconfig-gc is a Claude Code skill that performs garbage collection on a Claude Code configuration. It scans ~/.claude across eight channels (skills, memory, hooks, permissions, MCP servers, reminders, project history, caches) for redundant, stale, orphaned, or low-value items. A developer uses it when their .claude setup is bloated, session startup is slow, or a periodic config review is due. It matters because append-only configs rot silently, and this skill reclaims space with a forced human-in-the-loop, soft-delete-first, logged cleanup.1kinstalls 542Tastetaste is a Claude Code skill that adds a creative-direction layer to AI-generated music videos and short-form edits in the angelcore / cloud-trance / hyperpop visual family. It distills a named-genre aesthetic vocabulary (a ~70-entry genre taxonomy), a concrete mood, color, and light system, and a beat-synced editing grammar, then chains ECC's video skills into one production pipeline. Developers use it when the work is not just making a video render but making it feel intentional, such as building a music video, fancam, visualizer, or moodboard-driven reel. It matters because it decides the visual direction before the first generation instead of after.1kinstalls 543Agent Self Evaluationagent-self-evaluation is a Claude Code skill that has an agent pause after a non-trivial task and rate its own output against a structured 5-axis rubric. It scores accuracy, completeness, clarity, actionability, and conciseness on a 1-5 scale, requiring concrete evidence for any score below 5. A developer uses it after multi-file code changes, multi-step workflows, or debugging sessions to catch omissions before the user does. It is a reflection step, not a pass/fail gate.987installs 544Brand Discoverybrand-discovery is a Claude Code skill that runs a structured, adaptive brand identity interview across multiple sessions. It covers purpose, positioning, audience, personality, voice, narrative, and founder-brand tension in eight modules using laddering, 5 Whys, and projective techniques. A founder or brand strategist uses it to make a scattered or founder-dependent identity explicit. It matters because it produces a resumable, disk-persisted process and a master brandbook (90_SYNTHESIS.md) to brief designers, writers, and collaborators.985installs 545Benchmark Methodologybenchmark-methodology is a Claude Code skill that turns a scoped competitor set into comparable, evidence-anchored scores. It assesses each competitor on the same nine weighted dimensions (positioning, voice, visual craft, offer packaging, evidence, enterprise-readiness, thought leadership, pricing, and the client's strategic tension) with explicit 1-5 rubrics. A strategist uses it after competitive-platform-analysis has produced a tiered competitor set and before competitive-report-structure. It matters because consistent, defensible scores replace gut-feel rankings.979installs 546Competitive Platform Analysiscompetitive-platform-analysis is a Claude Code skill that scopes a competitive landscape before any benchmarking begins. It decides who counts as a competitor, which tier they belong to, and which sources to mine, working from the client's positioning brief. A strategist uses it as the first step of a three-skill competitive pipeline, before benchmark-methodology. It matters because the wrong competitor set makes the client look either unbeatable or doomed.978installs 547Competitive Report Structurecompetitive-report-structure is a Claude Code skill that assembles scored competitor profile cards into a decision-grade report. The report answers who the client competes with, how it competes, and where its defensible white-space is, built around the client's strategic tension. A strategist uses it as the final step of a three-skill competitive pipeline, after benchmark-methodology. It matters because it turns benchmark scores into a report that drives decisions rather than merely documenting the landscape.978installs 548Vue Patternsvue-patterns is a Claude Code skill that provides Vue.js 3 Composition API patterns and best practices. It covers component architecture, reactivity, composables, Pinia state management, Vue Router navigation, testing, and Nuxt SSR patterns, with a recommended feature-first project structure and script-setup component ordering. Developers use it when working on Vue, Nuxt, Vite, or Pinia projects, or when reviewing Vue single-file components. It matters because it standardizes modern Vue idioms like typed props, composables over mixins, and Pinia Setup Stores.965installs 549Ml Adoption Playbookml-adoption-playbook is a Claude Code skill giving an end-to-end methodology for adding machine learning to an existing non-ML codebase. It structures the work into five phases: problem framing and feasibility, data readiness and leakage prevention, architectural decoupling behind an API or service boundary, baseline model implementation and training, and handoff to MLOps. A developer uses it when asked to add a model such as recommendation, classification, or forecasting to a traditional software project without tightly coupling inference to core logic.954installs 550Golang PatternsThis skill teaches Go-specific design patterns including functional options, small interfaces, dependency injection, concurrency patterns (worker pools, context propagation), error handling (wrapping, custom errors, sentinels), package organization (cmd/internal/pkg structure), and testing strategies (table-driven tests, test helpers). Developers use it when designing Go APIs and packages, implementing concurrent systems, structuring projects, or refactoring codebases to follow idiomatic Go conventions. Key workflows cover constructor patterns for flexible configuration, defining interfaces at point of use for loose coupling, explicit dependency injection, safe concurrency with sync.WaitGroup and channels, and structured error handling with fmt.Errorf wrapping.0installs 551Security ReviewSecurity Review provides a comprehensive checklist and patterns for identifying and fixing common vulnerabilities across 10 critical areas: secrets management, input validation, SQL injection prevention, authentication/authorization, XSS, CSRF, rate limiting, sensitive data exposure, blockchain security, and dependency management. Developers activate it when implementing authentication, handling user input, creating API endpoints, working with secrets, or integrating payments. The skill covers concrete patterns (parameterized queries, httpOnly cookies, Zod validation, CSP headers, RLS policies) and pre-deployment verification steps, with examples for Next.js, Supabase, and Node.js stacks.0installs

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.