claude-code-skills

Use this skill when the user wants to contribute to an open-source GitHub repository — finding good first issues, claiming an issue, opening a pull request that matches the project's contributor guide, running the project's own verification stack locally, and shepherding the PR through CI and review. Trigger on phrases like "contribute to <repo>", "open source contribution", "find good first issues in <repo>", "open a PR for issue

Production-grade code quality enforcement for TypeScript, Next.js, Python, and FastAPI. This skill MUST be loaded when ANY .ts, .tsx, .py, .jsx, or .js file is being written, edited, reviewed, or created. Also trigger when the user asks to "create a component", "build a page", "add a feature", "write a function", "create an endpoint", "add a route", "implement a service", "review code", "refactor", "clean up", "improve code quality", "fix code", or any task that produces code output. This skill prevents the most common Claude failures: dumping all code into a single file, using function keyword instead of arrow functions, skipping error handling, missing type safety, writing insecure code, and creating untestable monolithic functions. Load this skill BEFORE writing any code — it governs structure, not just syntax.

This skill MUST be loaded when ANY file in an app/ directory is being read, reviewed, edited, or created - including page.tsx, layout.tsx, loading.tsx, error.tsx, not-found.tsx, route.ts, template.tsx, default.tsx, or any .tsx/.ts file using App Router patterns. Also use when the user asks to "read a page.tsx", "review a Next.js file", "edit a layout", "create a Next.js page", "add a route", "create a layout", "implement server components", "add client components", "create API routes", "implement server actions", "add loading states", "handle errors in Next.js", "create dynamic routes", "implement parallel routes", "add intercepting routes", or mentions Next.js App Router, React Server Components (RSC), 'use client', 'use server', next/navigation, or app directory patterns.

Optimizes GitHub repositories for maximum discoverability, professional presentation, and portfolio impact. Use this skill EVERY TIME you create a new repository, write a README, set up a GitHub project, prepare a portfolio project for deployment, or when the user mentions 'README', 'GitHub repo', 'push to GitHub', 'make it public', 'portfolio', 'open source', 'repo setup', 'GitHub profile', or 'publish project'. This skill covers repository naming, description optimization, topic tags, README structure, social preview images, badges, profile SEO, Google indexing, and promotion strategies. Trigger even when the user just says 'push this to GitHub' or 'set up the repo' — those are the moments where SEO matters most and is most often forgotten.

Converts design screenshots, Figma exports, and UI mockups into pixel-accurate React + Tailwind CSS + Shadcn/UI code. Use this skill EVERY TIME the user provides a screenshot, image, mockup, design, Figma frame, or any visual reference and asks to convert it to code, recreate it, build it, replicate it, or implement it. Also trigger when the user says 'make this', 'code this', 'convert this', 'build this design', 'implement this UI', or 'recreate this page' while providing an image. This skill forces a systematic, layer-by-layer analysis of the design before writing any code — preventing Claude from guessing, interpreting, or skipping visual details. If a screenshot is attached and the user wants code, USE THIS SKILL.

UI/UX engineering best practices for building production-quality frontend components with Tailwind CSS and Shadcn/UI. Use this skill EVERY TIME you build any React component, page, layout, dashboard, form, modal, table, sidebar, card, or any visual interface — even if the user doesn't explicitly mention UI/UX. This skill ensures Claude never ships components with missing states, poor accessibility, broken responsive layouts, or amateur UX patterns. Trigger on: any frontend component creation, any page layout, any form building, any dashboard, any interactive widget, any landing page, any data display, or when the user says 'build', 'create', 'make', or 'design' followed by any UI element. Also trigger when fixing, improving, or reviewing existing components. This skill assumes Tailwind CSS and Shadcn/UI as the styling stack.

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

This skill MUST be loaded when ANY Dockerfile, docker-compose.yml, docker-compose.override.yml, .dockerignore, or Docker-related configuration file is being read, reviewed, edited, or created. Also use when the user asks to "create a Dockerfile", "write a multi-stage build", "add a .dockerignore", "set up docker-compose", "containerize an application", "optimize a Docker image", "reduce Docker image size", "fix Docker build issues", "add health checks to containers", "set up a local dev environment with Docker", "create a multi-service setup", "debug container startup", "configure Docker volumes", "set up database containers", or mentions Dockerfile, Docker Compose, multi-stage builds, container images, build context, layer caching, ENTRYPOINT, CMD, docker-compose.yml, Docker health checks, or container orchestration for local development.

This skill MUST be loaded when ANY Kubernetes manifest file (.yml/.yaml) under a k8s/ directory is being read, reviewed, edited, or created — including Deployments, Services, ConfigMaps, Secrets, PVCs, HPAs, Namespaces, StatefulSets, Ingresses, or any resource manifest. Also use when the user asks to "create a Kubernetes deployment", "write K8s manifests", "add health probes", "configure liveness and readiness probes", "create a PersistentVolumeClaim", "set up a ConfigMap or Secret", "add an HPA", "create a K8s namespace", "write a StatefulSet", "expose a service with NodePort or ClusterIP", "set resource limits and requests", "add init containers", "configure envFrom with ConfigMap and Secret", "create a horizontal pod autoscaler", "debug pod startup issues", "fix CrashLoopBackOff", "troubleshoot ImagePullBackOff", "set up persistent storage for databases", "configure Redis persistence in K8s", "deploy a PostgreSQL StatefulSet", "add rolling update strategy", or mentions Kubernetes, K8s, kubectl, pods, deplo

This skill MUST be loaded when ANY .py file involving FastAPI is being read, reviewed, edited, or created. Also use when the user asks to "create a FastAPI app", "add an API endpoint", "create a REST API", "implement authentication", "add OAuth2", "create Pydantic models", "add request validation", "implement dependency injection", "create API routers", "add middleware", "handle errors in FastAPI", "create background tasks", "add WebSocket endpoints", "implement CRUD operations", "review a FastAPI route", "read a FastAPI file", or mentions FastAPI, Pydantic with FastAPI, async Python APIs, Starlette, or FastAPI patterns.

This skill MUST be loaded when ANY .py file is being read, reviewed, edited, or created that does NOT involve FastAPI, FastMCP, OpenAI Agents SDK, or n8n Code nodes. Also use when the user asks to "create a Python project", "set up a Python package", "write a Python script", "add type hints", "create a dataclass", "write pytest tests", "add logging", "create a decorator", "write a generator", "implement a context manager", "handle exceptions", "read/write files in Python", "use async/await in Python", "manage Python dependencies", "set up pyproject.toml", "create a virtual environment", "use pathlib", "parse arguments with argparse", or mentions Python typing, dataclasses, pytest, logging, PEP 8, Python packaging, pip, uv, poetry, or general Python development patterns.

This skill should be used when the user asks to "create a FastAPI app", "add an API endpoint", "create a REST API", "implement authentication", "add OAuth2", "create Pydantic models", "add request validation", "implement dependency injection", "create API routers", "add middleware", "handle errors in FastAPI", "create background tasks", "add WebSocket endpoints", "implement CRUD operations", or mentions FastAPI, Pydantic, async Python APIs, or Starlette.

This skill should be used when the user asks to "create an MCP server", "build MCP tools", "add MCP resources", "create MCP prompts", "implement MCP protocol", "deploy MCP server", "create AI tools", "expose API as MCP", "build Claude tools", "create LLM tools", or mentions FastMCP, MCP server, Model Context Protocol, MCP tools, MCP resources, MCP prompts, or server composition.

This skill should be used internally to decide when to use curated skills versus Context7 MCP for documentation lookups. Use this when handling technical questions about frameworks, libraries, or APIs to determine the optimal knowledge retrieval strategy.

This skill should be used when the user asks to "create an AI agent", "build a multi-agent system", "implement agent handoffs", "add guardrails to agents", "create function tools for agents", "implement agent tracing", "build agentic workflows", "create a triage agent", "implement input/output guardrails", "add tools to agents", or mentions OpenAI Agents SDK, agents, handoffs, guardrails, Runner, or multi-agent orchestration.