Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →

siviter-xyz/dot-agent

15 skills6.4k installs300 starsGitHub

Install

npx skills add https://github.com/siviter-xyz/dot-agent

Skills in this repo

1Create SkillCreate Skill is a meta guide for authoring effective agent skills that extend capabilities with specialized knowledge, workflows, and tool integrations. Skills are modular packages providing workflows, tool instructions, domain expertise, and bundled scripts or references. Progressive disclosure is mandatory: SKILL.md must stay under 200 lines with overflow in references/ files, yielding roughly 85 percent lower initial context load and activation under 100ms versus 500ms plus. The three-level loading model keeps metadata always in context, SKILL.md body on trigger, and bundled resources on demand. Structure requires YAML frontmatter with name and description plus markdown instructions, optionally scripts/, references/, and assets/ directories. Core principles demand concision because the context window is shared, appropriate degrees of freedom from high text guidance to low-parameter scripts for fragile ops, and testing across all target models since effectiveness depends on the underlying model. The 200-line rule is critical for maintainability and performance. Reference files progressive-disclosure.md, skill-structure.md, examples.md, and best-practices.md hold extended guidanc.2.7kinstalls2Cursor Best Practicescursor-best-practices is a siviter-xyz/dot-agent skill summarizing how to work productively in Cursor without typical AI coding mistakes. Core principles include starting complex tasks in Plan Mode (Shift+Tab), managing conversation context by letting the agent find files and starting fresh chats when scope shifts, extending agents through Rules for static context and Skills for dynamic capabilities, and reviewing AI-generated code with the same rigor as human contributions. Documented workflows cover test-driven development loops—writing tests first and iterating until green—and structured planning before large edits. Developers reach for cursor-best-practices when onboarding to Cursor, tuning agent behavior, or reducing wasted tokens from poor context hygiene. The skill is editor-meta guidance rather than language-specific implementation, applicable across SaaS, API, and CLI repositories using Cursor as the primary agent interface.620installs3Software Engineersoftware-engineer is a Claude agent skill from siviter-xyz/dot-agent that bakes software engineering principles into every coding session. The skill instructs agents to keep code self-documenting, avoid unnecessary comments, and leave committed code free of stray TODO and FIXME markers unless explicitly tracked. It applies when editing code, working inside software repositories, or performing general development tasks where style drift is common. Developers reach for software-engineer when they want an AI pair programmer that defaults to disciplined patterns instead of verbose or sloppy output. The skill covers code style, documentation habits, and workflow preferences in one reusable session profile.549installs4Pythonpython is an agent skill published in the siviter-xyz/dot-agent repository that steers Claude Code, Cursor, and compatible assistants toward idiomatic Python development workflows. The skill activates when agents encounter Python-specific tasks such as writing modules, refactoring functions, debugging tracebacks, or scaffolding backend scripts without switching to generic language-agnostic prompts. Because dot-agent packages domain skills as installable SKILL.md instructions, python injects Python conventions, standard library usage patterns, and project-context awareness into agent turns during build work. Developers reach for python when an agent session needs explicit Python guidance—virtual environment assumptions, typing hints, package layout, or test-friendly module structure—rather than ad hoc generic coding advice. The skill complements other dot-agent language and framework skills and fits early-through-mid implementation phases where Python source files, requirements manifests, and CLI entry points are being created or modified.332installs5Backend Engineerbackend-engineer is a progressive-disclosure skill from siviter-xyz/dot-agent (version 1.0.0) for building production-ready backend systems. The SKILL.md stays under 200 lines and routes to nine on-demand reference files covering technologies, API design, security, authentication, performance, architecture, testing, DevOps, and a unified implementation workflow. It includes a quick decision matrix mapping needs to stacks—NestJS for fast development, FastAPI for data/ML, Gin for concurrency, Axum for performance—and checklists for API, database, security, testing, and deployment phases. Best practices cite Argon2id passwords, Redis caching for 90% DB load reduction, a 70-20-10 test pyramid, and blue-green or canary Kubernetes deploys. Developers invoke backend-engineer when designing REST, GraphQL, or gRPC APIs, hardening auth, optimizing queries, or standing up microservices with Prometheus monitoring.310installs6Frontend Engineerfrontend-engineer is a dot-agent skill that gives coding agents a comprehensive React/TypeScript playbook emphasizing Suspense-based data fetching, lazy-loaded routes, and strict feature-folder organization. The skill documents eight core principles: lazy-load heavy routes and DataGrids, use SuspenseLoader instead of early-return spinners, prefer useSuspenseQuery for new data fetching, organize features under api/, components/, hooks/, helpers/, and types/ subdirectories, split styles at the 100-line threshold, and enforce import aliases (@/, ~types, ~components, ~features) with TypeScript strict mode and no any. Developers reach for frontend-engineer when creating pages, wiring TanStack Router routes, styling MUI v7 components, or optimizing bundle size in a React SPA. The dot-agent repo keeps SKILL.md under 200 lines with deeper references loaded on demand. Install with npx skills add siviter-xyz/dot-agent --skill frontend-engineer for Cursor, Claude Code, Codex, and other Agent Skills–compatible harnesses.290installs7Create Agents Mdcreate-agents-md is a siviter-xyz/dot-agent skill that guides creation of AGENTS.md files for small, project-specific instructions committed inside repositories. It supports a root AGENTS.md for whole-project commands, tech stack, testing, code style, architecture, and safety boundaries, plus nested subdirectory AGENTS.md files for monorepo packages, services, or test trees where local rules diverge. Developers reach for create-agents-md when adding folder-scoped agent rules that should live in git rather than reusable cross-project skills. The skill recommends short concrete references, exact CLI invocations, links to canonical example files, and keeping files focused instead of long prose documentation.285installs8Typescripttypescript: A skill for development. This provides functionality for development workflows.198installs9Semantic Gitsemantic-git: A skill for development. This provides functionality for development workflows.181installs10Cli Buildingcli-building: A skill for development. This provides functionality for development workflows.179installs11Context Engineeringcontext-engineering: A skill for development. This provides functionality for development workflows.176installs12Debuggingdebugging: A skill for development. This provides functionality for development workflows.131installs13Media Processingmedia-processing: A skill for development. This provides functionality for development workflows.131installs14Code Reviewcode-review: A skill for development. This provides functionality for development workflows.129installs15Psipsi: A skill for development. This provides functionality for development workflows.125installs

This week in AI coding

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

unsubscribe anytime.