Agent skills

Skills you can use with AI coding agents, indexed from public GitHub repositories.

Latest Most Popular

OpenCode SDK Development This skill should be used when the user asks to "create an OpenCode tool", "build an OpenCode plugin", "write a custom tool for OpenCode", "use @opencode-ai/sdk", "use @opencode-ai/plugin", "integrate with OpenCode", "create OpenCode hooks", "define tool schema", "use tool.schema", "work with OpenCode sessions", or needs guidance on OpenCode SDK patterns, plugin development, or custom tool creation.
majiayu000/claude-skill-registry 163
weavedb Comprehensive assistance with weavedb
majiayu000/claude-skill-registry 163
beam-pipeline-builder Beam Pipeline Builder - Auto-activating skill for Data Pipelines. Triggers on: beam pipeline builder, beam pipeline builder Part of the Data Pipelines skill category.
majiayu000/claude-skill-registry 163
amnesiac Autonomous software development for agents with no persistent memory. Use when building, testing, or maintaining code projects. Ensures all work is independently verifiable without context from previous sessions.
majiayu000/claude-skill-registry 163
planning-framework
majiayu000/claude-skill-registry 163
moai-cc-commands Designing Slash Commands for Claude Code. Create and optimize slash commands with proper argument parsing, tool permissions, and agent orchestration. Use when building workflow entry points, automation commands, or user-facing shortcuts.
majiayu000/claude-skill-registry 163
cloudflare-r2 Cloudflare R2 object storage - upload, download, list, delete files, presigned URLs. Trigger: When working with R2 storage, file uploads, object storage, presigned URLs.
majiayu000/claude-skill-registry 163
typescript-reviewer WHEN: TypeScript code review, type safety audit, tsconfig analysis, TS migration review WHAT: Type safety checks + any usage audit + generic patterns + strict mode + compiler options analysis WHEN NOT: React specific → nextjs-reviewer, Node.js backend → nodejs-reviewer, General code → code-reviewer
majiayu000/claude-skill-registry 163
vercel-react-best-practices React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
majiayu000/claude-skill-registry 163
maestro-delegation-advisor Expert system for AgentMaestro that helps decide when and how to delegate tasks to specialized AI agents (Claude, Codex, Gemini). Use when you need to determine which agent is best suited for a task, or when a task should be broken down and delegated to multiple agents.
majiayu000/claude-skill-registry 163
view-manual Use this skill to view the manual page for a given Tcl command.
majiayu000/claude-skill-registry 163
concurrency-patterns Implement thread-safe code, mutexes, semaphores, async/await patterns, and concurrent data structures. Use when handling parallel operations, race conditions, or building high-performance concurrent systems.
majiayu000/claude-skill-registry 163
yaml-master PROACTIVE YAML INTELLIGENCE: Automatically activates when working with YAML files, configuration management, CI/CD pipelines, Kubernetes manifests, Docker Compose, or any YAML-based workflows. Provides intelligent validation, schema inference, linting, format conversion (JSON/TOML/XML), and structural transformations with deep understanding of YAML specifications and common anti-patterns.
majiayu000/claude-skill-registry 163
objection-pattern-detector Mine lost deal notes to identify recurring objection patterns. Create objection response playbooks from won deals.
majiayu000/claude-skill-registry 163
url-readmode-markdown Convert a web URL into reader-mode Markdown and save it as a .md file named after the content's topic/title. Use for tasks like extracting article text, cleaning noisy pages, and producing a readable Markdown file from a URL.
majiayu000/claude-skill-registry 163
continental-critical Master Continental philosophy and Critical Theory. Use for: post-structuralism, deconstruction, Frankfurt School, hermeneutics, psychoanalysis. Triggers: 'Foucault', 'power/knowledge', 'deconstruction', 'Derrida', 'différance', 'critical theory', 'Frankfurt School', 'Adorno', 'Habermas', 'genealogy', 'discourse', 'Lacan', 'Deleuze', 'rhizome', 'biopolitics', 'ideology', 'alienation', 'reification', 'hermeneutics', 'Gadamer', 'post-structuralism', 'logocentrism'.
majiayu000/claude-skill-registry 163
Global Commenting Write meaningful code comments that explain WHY rather than WHAT, focusing on business logic, non-obvious solutions, workarounds, and complex algorithms while keeping code self-documenting. Use this skill when adding comments to explain rationale, documenting complex business logic, explaining workarounds or temporary solutions, describing performance optimizations, writing function documentation (JSDoc, docstrings, XML docs), or reviewing code for appropriate commenting. Apply when working on any code file that contains logic requiring explanation, public API functions, complex algorithms, security-critical code, or architectural decisions. This skill ensures comments explain rationale not implementation (WHY not WHAT), self-documenting code through clear naming (refactor unclear code instead of commenting), concise and evergreen comments (no who/when dated comments - Git tracks this), links to external resources for context, proper function documentation format (JSDoc for TS/JS, docstrings for Python, XML docs for .NET), no TODO/FIXME in committed code (use issue tracking system instead and reference ticket), and deletion of dead commented-out code (Git history preserves it).
majiayu000/claude-skill-registry 163
text-summary 将长文本压缩为简洁摘要，支持多种语言
majiayu000/claude-skill-registry 163
mckinsey-consultant McKinsey顾问式问题解决系统。从商业问题出发,通过假设驱动的结构化分析方法,生成McKinsey风格研究报告和PPT。融合Problem Solving方法论、MECE原则、Issue Tree拆解、Hypotheses形成、Dummy Page设计、智能数据收集和专业PPT生成能力。
majiayu000/claude-skill-registry 163
google-drive-manager Expert in Google Drive file management. Use when searching, copying, moving, uploading, downloading, sharing, or managing files and folders in Google Drive.
majiayu000/claude-skill-registry 163
model-usage Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.
majiayu000/claude-skill-registry 163
docs-crackmac-alexa-skills-kit
majiayu000/claude-skill-registry 163
debug-surefire Debug Maven Surefire unit tests by running them in JDWP "wait for debugger" mode (`-Dmaven.surefire.debug`) and attaching to the forked test JVM using **jdb** (preferred for CLI/agent debugging), IntelliJ, or VS Code. Use when asked to debug/step through a failing JUnit test, attach a debugger to a Maven test run, or run `mvn test -Dtest=Class[#method]` suspended on a port (including multi-module `-pl` runs). The JVM will block at startup until a debugger attaches; the agent should attach with `jdb -attach <host>:<port>` and drive the session from the terminal.
majiayu000/claude-skill-registry 163
enforce-file-limits Enforce 500-line file size limits. Warn at 400 lines. Guide splitting large files into focused modules. Apply when writing or editing code.
majiayu000/claude-skill-registry 163