Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Orchestrate the 8-layer SDD workflow - recommend the right skill for an intent, detect the current position and next steps, explain the BRD→…→Code flow, and enforce the upstream-artifact policy. Use when unsure what to do or which skill to run next.
Plan a large feature as a ROADMAP.rev1 written with the user — you are the product manager first and the senior software architect second, never the coder. Use when the user asks to plan a large feature, write a roadmap, or invokes /draft-roadmap <feature>. Settles the product shape — the one sentence, and the ladder of use cases smallest to largest — then derives a roadmap of small, verifiable, delegatable milestones and plans from it, ready for review and for parallel execution
[Beta] Creates permanent eval cases from promoted learnings and runs regression checks against them. Turns failures into test cases that prevent silent regression. This is the outer loop's regress-test step. Use when a learning is promoted and has a clear pass/fail condition, or on cadence to verify promoted rules still hold.
Primary pipeline execution step after /prd-to-issues or for clearly scoped implementation work. Use to build, verify, and commit a concrete slice, delegating to /tdd for backend work and behavior-heavy frontend logic when red-green-refactor will reduce risk. Not for shaping or pre-merge review.
Execute a signed-off PLAN.final.md for a single feature yourself, phase by phase, on the current branch — you are the engineer, not a delivery lead. Use when the user invokes /execute-plan, or asks to implement, build or continue an approved plan that is small enough for one person, e.g. "/execute-plan spell-check", "/execute-plan spell-check p1-p3", "build the rest of the plan while I sleep". Resolves the requested phases against the final doc, builds each as its own verified increment, runs its verification before the next begins, and commits each landed phase without waiting for approval. Questions that need the user are recorded with a recommendation instead of blocking. For a roadmap large enough to want parallel sub-agents, use execute-roadmap instead
Execute an approved roadmap or plan by orchestrating sub-agents — in parallel, in isolated checkouts, where the runtime allows — you are the delivery lead, not the coder. Use when the user invokes /execute-roadmap, or asks to implement, execute, or oversee an approved roadmap or plan after draft-roadmap or build-feature has produced a signed-off final doc, e.g. "/execute-roadmap implement m2, plans 3.1-3.4, parallelize" or "oversee the roadmap until it is complete". Resolves the requested scope against the final doc, gives each plan to its own sub-agent, reviews and verifies each result against the project's principles, then merges, cleans up, and commits without waiting for approval. Honors model/effort directives where the agent tool supports them; questions that need the user are recorded with a recommendation instead of blocking.
Google Gemini CLI orchestration for AI-assisted development. Capabilities: second opinion/cross-validation, real-time web search (Google Search), codebase architecture analysis, parallel code generation, code review from different perspective. Actions: query, search, analyze, generate, review with Gemini. Keywords: Gemini CLI, second opinion, cross-validation, Google Search, web research, current information, parallel AI, code review, architecture analysis, gemini prompt, AI comparison, real-time search, alternative perspective, code generation. Use when: needing second AI opinion, searching current web information, analyzing codebase architecture, generating code in parallel, getting alternative code review, researching current events/docs.
Chat with Grok Bot teammates, message a named Grok Bot, create a new Grok Bot, list Grok Bots, read a Grok Bot transcript, or operate the xAI Grok Bot desktop app from Cursor, Claude Code, or Codex instead of the UI. Use when the user mentions Grok Bot, Grok teammates, Reed, Agent Computer, spinning up a new bot, or talking to Grok from the terminal.
Cheat sheet + workflow for launching interactive coding-agent CLIs (Claude Code, Gemini CLI, Codex CLI, Cursor CLI, and pi itself) via the interactive_shell overlay or headless dispatch. Use for TUI agents and long-running processes that need supervision, fire-and-forget delegation, or headless background execution. Regular bash commands should use the bash tool instead.
Full health check of all your skills in one report. Use when the user wants to check for errors, find duplicates, detect broken skills, or get a complete overview of skill health. Pass --brief for inventory only. Trigger with '/janitor-report'.
Use when the user wants to persist a single insight from the current chat as a structured, project-local doc — debugging breakthroughs, hard-won config, workflow steps, or post-mortem lessons. Triggers: 'save this', 'document this', 'we figured it out', '记录下来', '总结一下'. Writes to <project>/.{trae,claude,cursor,windsurf}/knowledges/. For global compounding knowledge (architecture, patterns, APIs), use llm-wiki instead.
Build multi-agent AI systems with LangGraph — the low-level orchestration framework for stateful, graph-based agent workflows. Covers supervisor, swarm, and hierarchical multi-agent patterns; subgraph composition; state management (checkpointers/stores); persistence; evals; and production debugging. Reach for this when designing agent architectures that need cycles, conditional branching, parallel execution, or human-in-the-loop patterns. Do not use this skill for unrelated requests; route to the nearest named specialist.
Use PUA talk to motivate OpenClaw (lobster) to work more efficiently and with better quality. Triggered when you need to push an AI to complete tasks more efficiently and with higher quality, applicable to: (1) low task execution efficiency (2) output quality not good enough (3) needing the AI to focus on the work (4) tasks that need to be done in one go without dawdling
Look up current API signatures from installed Mastra packages in node_modules/@mastra/*/dist/docs/. Use this skill to verify exact parameters, constructor arguments, and function signatures that match the installed version. Mastra APIs evolve rapidly - embedded docs ensure accuracy. Covers: Agent constructor parameters, Workflow API, Tool API, Memory configuration, Storage options, and all @mastra/* package exports. Returns actual TypeScript definitions and JSDoc from installed packages. Always use for API verification questions to prevent outdated answers.
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
Canonical Claude Code authoring kit covering Skills, sub-agents, plugins, slash commands, hooks, memory, settings, sandboxing, headless mode, and advanced agent patterns. Use when creating Claude Code extensions or configuring Claude Code features.
Main entry point for the Show Me The Money business automation suite. Dual mode: pre-task routing (which skill fits your problem) + post-task navigation (a skill just finished — what's next). Use when the user wants to start a business, automate operations, generate revenue, find product ideas, set up marketing, or run any business function autonomously. Also use when the user says 'show me the money', '来财' (the suite's Chinese name), 'make money', 'start a business', 'automate my business', 'build a company', 'what's next', or '下一步怎么走'. Includes a beginner tutorial mode triggered by '/money onboarding', '/money tutorial', '/money 新手', '/money 教程', 'how do I use this', or '这个怎么用' — a zero-input guided walkthrough that produces a demo business brief.
Run Claude Code or Codex CLI on Muapi (api.muapi.ai) using Muapi credits instead of a separate vendor account. Use when the user wants a local coding agent to use Muapi as its model provider, asks which Muapi models work with Claude Code or Codex, or needs the base URL, API key variable, provider block or model name for either agent. Covers Claude Code (Anthropic Messages) and Codex CLI (OpenAI Responses), model discovery, verification and troubleshooting.
Use when asked to fix, build, refactor, review, verify, or release software and the work is non-trivial — including delivering a change request through planning, gates, implementation, review, verified PR, and authorized release; continuing work started elsewhere with no packet via evidence-based position assessment; or operating tracked work items in Linear, Jira, or Notion via their tooling skills. neckbeard routes the change through framing, discovery, design, implementation, review, verification, and learning — choosing the smallest *safe* intervention, proving it at the real delivery boundary, and leaving an inspectable evidence ledger. For issue-to-PR work, conditionally loads a 9-phase journey with gates and lifecycle integration. Composes specialist catalog skills rather than replacing them. Not a persona or LOC-minimizer. Do not use this skill for unrelated requests; route to the nearest named specialist.
Proactively convert approved OpenSpec specs into Beads issues when the user applies a change or explicitly approves implementation. Creates trackable work with dependencies, discovers gaps, and maintains an audit trail between planning and execution.
Triggers when the user says "opsx batch fix", "batch opsx-jira-fix", "opsx-jira-fix-batch", "batch OpenSpec Jira fix". Thin trigger shell: parse Jira IDs/URLs and enqueue them into goal-driven-batch with Engine opsx-jira-fix-workflow. Does not run fixes or merge. Do NOT use for a single issue (opsx-jira-fix-workflow) or a non-Jira queue (goal-driven-batch).
Operate Orca-managed worktrees, folder contexts, terminals, repos, automations, artifacts, skill sharing, worktree comments, and Orca's embedded browser through the `orca-np` CLI. Use when the user says "$orca-cli", "Orca worktree", "child worktree", "spawn codex/claude in a worktree", "read/wait/send Orca terminal", "handoff" / "handover" / "give this to another agent", "Orca browser", "orca artifacts", or "share skills". Prefer it over raw git worktree, ad hoc PTYs, or Computer Use when Orca state is involved. Use Computer Use only when a visible window needs GUI control that a CLI, filesystem, or API cannot do
Set up the Peach harness system in the target project. Tidies CLAUDE.md into a minimal entry point and creates/updates AGENTS.md at 95–130 lines. Use with the keywords "하네스 설정", "프로젝트 초기 설정", "CLAUDE.md 정리", "AGENTS.md 업데이트", "세션 시작 설정".
Generates project context and codebase documentation for AI agents. Creates `.agents/` directory with project overview and deep codebase analysis (architecture, conventions, testing, integrations, checklist, workflows, features when vertical slicing detected, review). Use when starting work on a project, onboarding to an existing codebase, generating project documentation for agents, mapping codebase patterns, or incrementally refreshing the index from recent git changes. Triggers: "initialize .agents", "setup project index", "index project", "map codebase", "analyze codebase", "project overview", "codebase summary", "onboarding to this repo", "integrate feedback", "sync knowledge", "refresh codebase", "sync codebase", "patch index". Not for feature work, Obsidian session notes, or dev tooling setup like prettier or eslint.
Generates comprehensive planning documentation for web application projects, structuring work into context-safe phases with built-in verification criteria. Creates IMPLEMENTATION_PHASES.md, DATABASE_SCHEMA.md, API_ENDPOINTS.md, ARCHITECTURE.md, and other planning docs based on project needs. Optimized for Cloudflare Workers + Vite + React stack. Use when starting new projects, adding major features, or restructuring existing work into manageable phases. Keywords: project planning, planning documentation, IMPLEMENTATION_PHASES.md, DATABASE_SCHEMA.md, API_ENDPOINTS.md, ARCHITECTURE.md, UI_COMPONENTS.md, TESTING.md, AGENTS_CONFIG.md, phased development, context-safe phases, verification criteria, exit criteria, planning docs generator, web app planning, Cloudflare Workers planning, Vite React planning, project structure, project phases, major features planning, new project setup
Read a customer's live Langfuse traces, sessions, and scores via the Langfuse MCP server, pattern-match against a known-failure catalog, and write a findings report split into Ratel-flavored opportunities and stack-agnostic low-hanging fruit. Use to scan the dashboard, dig into failures, review recent sessions or the partner's traces, audit their observability, see what the agent's doing wrong, or `/ratel-langfuse-analyze`. Writes to <repo>/.ratel/ (accumulates); fails fast if MCP is down, honest-skips clean data, never edits code. Follow-up to /ratel-langfuse-integrate, reached after /ratel-observability-assessment picks Langfuse.
Use to resolve Blocked, blocked, or pending-decision items in a development plan; not to write a plan from scratch, review it overall, or implement it. Verify independently and confirm interactively; raise CRs for external dependencies through CRD, save decision records, and update the plan.
Pick up unfinished work from docs/<feature>/HANDOFF.md in a fresh session — you are the agent the handoff was written for, and you verify it before you trust it. Use when the user invokes /resume-handoff <feature>, points at a HANDOFF.md, or asks to continue, resume, or pick up where the last session stopped on a plan or roadmap built by execute-plan, execute-roadmap or build-feature, e.g. "pick up the handoff and keep going while I sleep". Reads the project's agent orientation and the handoff, reconciles its claims with git and the code, surveys the pending merge and proposes the scope — then stops and asks the user to confirm the merge and plan of work, or give new instructions, before anything changes the tree. On confirmation it deletes the consumed handoff, merges through the merge-branch gate, continues under execute-plan or execute-roadmap, and writes a fresh handoff at the end if work is still owed. For writing a handoff, use write-handoff; for retiring finished work, use close-plan or close-roadmap.
Use when orchestrating a code review workflow across multiple LLMs or agents, synthesizing review findings, producing checklist-driven fix scopes, applying selected fixes, running regression tests, independently verifying fixes, updating review status, or keeping code review process artifacts isolated under code_review/.
Self-improving skills toolkit that watches real agent sessions, detects missed triggers, grades execution quality, and improves skill packages through evals, replay, baselines, review, and post-deploy watch. Use when verifying or publishing a skill, improving instructions or routing, checking skill health, grading sessions, viewing the dashboard, ingesting agent histories, auditing installed skills, finding and temporarily using local skills for a task, managing the local Skill Library, scaffolding reusable workflow skills, or running autonomous improvement loops. Trigger whenever a user asks about skill performance, health, triggers, undertriggering, overtriggering, evolution, evaluation, or how their skills are doing, even if they do not name SelfTune explicitly.
Bring your own SIP trunk to Bolna (BYOT) for AI voice agents. Create a trunk with gateways, IP-based or user/password auth, UDP/TCP/TLS transport, optional SDES (SRTP) media encryption, add DID numbers, and wire them to agents for inbound or outbound calls. Use with Twilio Elastic SIP, Plivo Zentrunk, Telnyx, Vonage, Zadarma, DIDWW, or any standards-compliant trunk. Covers IP whitelisting (`13.200.45.61`), origination URI setup, and the classic SRTP-mismatch / UDP-fragmentation / TLS-without-SDES diagnostic flow.
Search and discover Agent Skills from the Skills Pool directory. Use when the user needs a new agent skill, wants to find tools for a specific task, or asks you to search for skills online. Not for local/already-installed skills.
A skill that keeps the takt-* skill group (takt-task-builder, takt-workflow-builder, takt-facet-builder, takt-analyzer, takt-optimizer) up to date with the latest takt version when nrslib/takt releases a new version. It detects diffs in the TypeScript schemas (taskRecordSchemas.ts / taskExecutionSchemas.ts), workflow YAML, and facet Markdown, and systematically updates SKILL.md and reference documents (task-schema.md, etc.). Triggers: "update takt skills", "check freshness of takt-* skills", "handle takt version upgrade", "check whether skills are outdated", "takt skill updater"
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.