akillness/skills-template
44 skills1.2k installs1.8k starsGitHub
Install
npx skills add https://github.com/akillness/skills-templateSkills in this repo
1Backend TestingThis skill turns backend test ambiguity into one practical test packet. A developer uses it for API, service, repository, and auth-flow coverage design, fixture and reset strategy, container-versus-mock dependency choices, and flaky-suite stabilization across local and CI. It picks the smallest credible layer mix and splits local, PR, and slower lanes.75installs2Data AnalysisData Analysis is a skill that analyzes datasets with a decision-first workflow. A developer uses it to inspect CSV, JSON, Parquet, or SQL extracts, validate data quality, explain trends, compare segments, and summarize experiment or telemetry results into evidence-backed findings. It frames the decision question, runs a data-quality trust check, chooses the cheapest analysis lane, and separates observation from interpretation.62installs3Web AccessibilityA routing-first skill for web accessibility remediation and verification. It classifies which accessibility surface is failing (semantics, keyboard/focus, labels/announcements, visual perception/reflow, media alternatives or routed-app navigation feedback) and turns vague WCAG, axe or Lighthouse requests into one remediation packet. It separates automated findings from manual and assistive-technology follow-up. Developers use it when a modal, form, nav or routed flow needs an accessibility fix plan before release.36installs4PlannotatorPlannotator is a skill that reviews and annotates AI agent plans or diffs in a visual browser UI, then approves, requests changes, or archives them. A developer uses it as a human approval gate before implementation or for visual diff review after coding. It supports Claude Code, Codex CLI, Gemini CLI, and OpenCode, and can save reviewed plans to Obsidian or Bear.35installs5Bmad GdsThis skill orchestrates AI-assisted indie game development across concepting, GDD shaping, milestone planning, and launch readiness. A solo developer or small team uses it to turn a mixed packet of idea notes, backlog state, playtest feedback, or a release target into the next game-production artifact. It routes specialist work to the right downstream game skill.32installs6Vibe KanbanA skill for running multiple AI coding agents (Claude, Codex, Gemini, OpenCode) on a visual Kanban board. Moving a card to In Progress automatically creates a git worktree and starts an agent, and results surface as diffs and logs for review. It runs Vibe Kanban as an MCP server exposing vk_ tools, and it integrates GitHub PR creation. Developers use it to break an epic into parallel agent tasks and track their status visually.32installs7Code RefactoringCode Refactoring is a skill that turns a cleanup request into one behavior-preserving refactor brief. A developer uses it to simplify a messy function or module, split an oversized diff into safer slices, freeze behavior before touching fragile legacy code, or plan a repeated codemod migration. It picks one refactor mode, makes the behavior guardrail explicit, and stages the work in reviewable slices.30installs8OpencontextOpenContext is a skill that manages active project and repo memory for coding agents. A developer uses it to load a compact read-first packet, search prior decisions, store reusable conclusions, or wire up the OpenContext CLI so agents do not start cold between sessions. It classifies the memory need, picks one packet, and routes wiki, graph, vault, or repo-packing requests to other tools.30installs9Workflow AutomationA skill for designing repo-scoped recurring workflow automation. It classifies the recurring workflow, chooses the right automation surface (npm scripts, Make, just, Task, hooks or a local-CI mirror), and keeps repo-local glue separate from environment provisioning and deployment. It makes automation idempotent and non-interactive. Developers use it to replace copy-paste shell rituals with stable command entrypoints for setup, build, test, lint and release prep.30installs10Marketing AutomationMarketing Automation is a skill that routes a marketing goal to one of 23 sub-skills across CRO, copywriting, SEO, ads/analytics, and growth. A developer uses it to generate a repeatable deliverable tuned to a single KPI, with a structured brief, validation checklist, and A/B test plan. It outputs assets like landing-page CRO changes, email sequences, or programmatic SEO templates.29installs11RalphRalph (Ouroboros) is a skill for specification-first AI development. A developer uses it to crystallize vague requirements into an immutable spec via Socratic interviewing before writing code, then execute and loop autonomously until verification passes. It provides ooo commands, Nine Minds lateral-thinking agents, drift detection, and a persistent Ralph loop with state files across Claude Code, Codex CLI, Gemini CLI, and OpenCode.29installs12Video ProductionA skill for planning automated and programmable video production across code-first, template-first and hybrid pipelines. It normalizes a production brief, picks one production mode (Remotion code-first, template/API automation, clip repurposing or manual-finish hybrid), and returns one implementation-ready packet with asset, QA and handoff guidance. Developers use it for repeatable video generation such as branded short-form clips, personalized videos or localized variants.29installs13Bmad IdeaThis skill is a pre-planning creative intake router that turns raw ideas into a single clear concept artifact. A developer or founder uses it when an early-stage product, GTM, content, or game idea needs framing before PRDs or sprint plans. It picks one framing mode (problem, audience/value, concept, game pillars, or story) and routes the next step to a neighboring skill.28installs14File OrganizationFile Organization is a skill that chooses and evolves project or repository structure without cargo-culting one folder tree. A developer uses it to decide between framework defaults, feature/domain folders, workspace or monorepo boundaries, docs taxonomy, or engine-aware game layouts, and to clean up a messy repo. It classifies the repo, picks one organization mode, defines boundary rules, and plans an incremental migration.28installs15Monitoring ObservabilityMonitoring & Observability is a skill that sets up metrics collection, log aggregation, alerting, and health checks for applications and infrastructure. A developer uses it before production deployment or when diagnosing bottlenecks and incidents. It provides Prometheus instrumentation, alert rules, Winston structured logging, Grafana dashboards, and health-check code, organized around the four golden signals.28installs16Changelog MaintenanceChangelog Maintenance is a skill that writes and maintains release-history artifacts such as CHANGELOG.md updates, release notes, migration and deprecation notices, and game patch notes. A developer uses it when the job is turning shipped evidence into the smallest truthful release-writing packet for developers, customers, internal stakeholders, or players. It classifies one release-writing mode per run and routes deeper work to neighboring documentation skills.27installs17Codebase SearchCodebase Search is a skill that routes repo-navigation requests to the right search approach before editing anything. A developer uses it to find where something is defined, who references it, which files own a config or content surface, and what to inspect before a change without drifting into debugging or refactoring. It normalizes the request into one search packet, narrows scope, and returns a compact evidence map.27installs18Database Schema DesignDatabase Schema Design is a skill that designs storage-model and migration-safety packets for relational, document-heavy, and hybrid data systems. A developer uses it to turn domain entities into tables, collections, ownership boundaries, and lifecycle rules, and to justify constraints, indexes, multi-tenant scope, and audit structures. It classifies one storage-design packet, gathers minimum evidence, and plans staged schema evolution.27installs19DebuggingDebugging is a skill that runs a reproduce, isolate, verify workflow for concrete bugs, regressions, flaky failures, and environment-specific behavior. A developer uses it when they already have a failing command, test, request, or UI flow and need root-cause diagnosis or fix verification. It freezes the failure definition, builds the cheapest reliable reproducer, isolates the failing boundary, and runs one discriminating check per hypothesis.27installs20Deployment AutomationDeployment Automation is a skill that plans and reviews release execution for web, backend, and fullstack systems. A developer uses it when the system can already build and the main job is shipping or recovering a release safely - choosing between preview, promotion, canary, and rollback paths, or tightening deploy gates around health checks and sign-off. It freezes the release profile, gathers the real deploy facts, and selects one rollout strategy.27installs21Presentation BuilderPresentation Builder is a skill that builds editable slide decks such as investor, roadmap, launch, architecture, and workshop decks. A developer uses it when the deliverable is a reviewable, presentable deck rather than prose. It chooses one deck mode, one artifact packet, and one handoff surface, builds an HTML-first deck with the slides-grab tool, then exports to PPTX, PDF, Google Slides, or Figma Slides.27installs22Responsive DesignResponsive Design is a skill that plans mobile-first, container-aware responsive layouts before patching breakpoints everywhere. A developer uses it to fix overflow, wrapping, layout collapse, and responsive media, or to decide what should adapt at the viewport level versus inside a component. It classifies the failure, picks the right adaptation surface, and produces a strategy packet with explicit zoom and reflow verification.27installs23Standup MeetingStandup Meeting is a skill that turns messy daily-sync requests into one coordination-cadence decision plus the lightest standup mode that still catches blockers and handoffs. A developer or team lead uses it to choose between live, async, hybrid, less-frequent, or no recurring standup, then run a bounded 10-to-15-minute sync. It outputs a Coordination Cadence Brief with facilitation flow, update templates, and blocker routing.27installs24Agent BrowserThis skill runs fresh-session browser automation for deterministic website verification. A developer uses it when they need a clean disposable browser, stable snapshot refs, and explicit before/after evidence across forms, page state, and accessibility-tree checks. It follows an observe-act-observe loop and supports named isolated sessions.26installs25Code ReviewCode Review is a skill that conducts thorough, constructive code reviews for quality and security. A developer uses it when reviewing pull requests, checking code quality, identifying bugs, or auditing security. It walks an eight-step process covering architecture, naming, SOLID principles, security vulnerabilities, performance, testing coverage, and documentation, and prioritizes findings by severity.26installs26Design SystemDesign System is a skill that produces UI designs using design tokens, layout rules, motion guidance, and accessibility checks for consistent frontend development. A developer uses it to generate production-quality UI from prompts, maintain a consistent visual language across screens, and build a systematic typography, layout, and motion guide. It defines tokens, generates component structure, and validates accessibility before handoff.26installs27Log AnalysisLog Analysis is a skill that inspects application and web-server logs to surface errors, slow requests, and anomalous access patterns. A developer uses it when debugging an issue, checking system health, or investigating an incident. It provides read-only grep and awk recipes for Apache, Nginx, application, and JSON log formats and a report template.26installs28Technical WritingA skill for writing internal technical documentation for engineers and operators. It covers technical specs, architecture docs, ADRs, runbooks, migration notes and internal implementation guides. It first classifies the document mode and audience, then drafts from a mode-specific skeleton. It routes published API portals, end-user tutorials and changelog work to separate skills.26installs29Testing StrategiesA skill for deciding test policy and release confidence rather than writing test code. It normalizes a change into a validation packet, names which gate is being decided (merge, release or scheduled), and picks the smallest convincing layer mix. It states flake and exception rules honestly and routes implementation, debugging and code review to other skills. Developers use it when 'add more tests' advice needs to become a concrete gate design.26installs30AgentationThis skill captures rendered-UI feedback and encodes it as structured packets with selectors, element paths, and bounding boxes. A developer uses it when a human reviewer needs to point at a real UI element and hand a precise fix packet to a coding agent. It supports copy-paste review, an MCP-backed synced watch loop, and self-driving critique modes.25installs31Authentication SetupThis skill designs or refactors product authentication setup for web apps and APIs. A developer uses it to choose an auth lane (hosted, framework-native, platform-native, enterprise add-on, or self-hosted), define session versus token boundaries, and wire OAuth, passkeys, org models, and SSO/SCIM. It records callback URLs, cookie domains, and migration boundaries before implementation.25installs32Api DesignThis skill designs or refactors REST and GraphQL API contracts before implementation. A developer uses it for resource modeling, endpoint and schema shape, naming, versioning, pagination, filtering, and error semantics. It produces a shared contract artifact (OpenAPI or GraphQL SDL) and surfaces compatibility tradeoffs.24installs33Api DocumentationThis skill publishes or refreshes developer-facing API documentation for REST, GraphQL, webhook, and SDK surfaces. A developer uses it when the contract already exists and the goal is helping integrators reach first success with truthful examples and error, retry, and limit guidance. It classifies each request into a single primary doc mode and manages large-surface navigation.24installs34Pattern DetectionPattern Detection is a skill that routes repeated-pattern, rule, and anomaly work into one detection packet before suggesting fixes. A developer uses it to get a reusable scan or shortlist of suspicious repeated shapes and outliers across code, logs, telemetry, and metric tables. It returns grouped candidates with confidence and false-positive notes, then routes diagnosis or remediation to other skills.24installs35Prompt RepetitionPrompt Repetition is a skill that decides when repeating a full prompt is a useful accuracy hack for lightweight or non-reasoning LLM tasks, and when the real fix is better prompt structure, retrieval, or a stronger model. A developer uses it for long-context retrieval, options-first multiple choice, or position-sensitive prompts on models like Haiku, Flash, or mini. It gives a bounded decision rule plus an input-token cost caveat.24installs36State ManagementA skill for implementing state management in React frontend applications. It helps a developer decide between local state (useState, useReducer) and global state (Context API, Redux, Zustand, Recoil) based on scope. It ships worked examples for an authentication context and a persisted Zustand shopping cart. Developers use it when props drilling or shared cross-component data becomes hard to manage.24installs37Vercel DeployA skill that deploys applications and websites to Vercel without requiring authentication. It packages the project into a tarball (excluding node_modules and .git), auto-detects the framework from package.json, and returns a live Preview URL plus a claimable deployment link. Developers use it to push an app live or create a preview deployment quickly.24installs38Git WorkflowGit Workflow is a skill that routes local Git work into the safest next move. A developer uses it to prepare a branch, clean up commits, choose between merge and rebase, resolve local conflicts, push rewritten history safely, or recover lost commits. It normalizes the Git intake, picks one workflow mode, and recommends the smallest reversible fix using a safety ladder.23installs39Performance OptimizationPerformance Optimization is a skill that improves application speed, efficiency, and scalability. A developer uses it to raise Lighthouse scores, shrink bundle size, speed up rendering, or fix slow database queries. It covers measurement (Lighthouse, Web Vitals), React memoization and code splitting, image optimization, and backend fixes like N+1 removal, indexes, and Redis caching.23installs40System Environment SetupA skill for building reproducible developer environments so a repo runs consistently across machines and containers. It covers runtime version pinning, Docker Compose, devcontainers, bootstrap scripts and onboarding flows. It routes narrow .env and app-config work to a separate environment-setup skill. Developers use it when a project is broken on a new machine or when local setup has drifted from the docs.22installs41Bmad OrchestratorThis skill orchestrates BMAD workflows for structured AI-driven development. A developer uses it to route work across Analysis, Planning, Solutioning, and Implementation phases, running TEA (Task-Execute-Architect) cycles within each phase. It adds automated architect validation and cross-phase traceability before opening a human review gate.21installs42Agentic WorkflowThis skill collects practical AI agent workflow patterns and productivity techniques for daily development. A developer uses it for agent commands, keyboard shortcuts, Git integration, MCP server usage, and session management. It also describes multi-agent orchestration and TDD workflow patterns across Claude Code, Gemini CLI, and Codex.19installs43Agent EvaluationThis skill designs and implements evaluation systems for AI agents. A developer uses it to build graders (code-based, model-based, human), pick benchmarks by agent type, and set up CI/CD eval pipelines. It follows an 8-step roadmap and covers coding, conversational, research, and computer-use agents.18installs44Image GenerationImage Generation is a skill that generates images via MCP using Gemini models or compatible services. A developer uses it to create marketing assets, UI placeholders and icons, and presentation visuals from structured prompts with defined aspect ratios and brand colors. It configures the MCP environment, structures the prompt, selects a model, generates and reviews variants, and records prompt metadata for reproducibility.17installs