Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Use when facing complex reasoning tasks — multi-step math, logic puzzles, decisions with tradeoffs, problems where direct answers fail, or when you need to show your work. Triggers on arithmetic errors, shallow analysis, or "I'm not sure" hedging.
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).
Spec-driven feature work with traceable requirements, design, tasks, and user acceptance testing (UAT). Use when planning, designing, implementing, or validating a feature from a story, feature PRD, or feature RFC. Not for unknown-bug diagnosis, standalone product or technical documents, PR or commit mechanics, or backlog tracking.
Cross-runtime playbook for multi-agent collaboration, agent teams, swarm orchestration, parallel task distribution, capability discovery, and quality gates on Claude Code, Codex, OpenClaw, and Cursor. Use for 多agent、agent协作、agent编排、并行agent、分工协作、拉团队、多代理协作 or complex work needing independent roles and review. Map the six-stage workflow to the active host's actual capabilities; report sequential degradation when parallel agents are unavailable
Structured clarification and requirements gathering through focused dialogue or with dry code. Use when a task is ambiguous, underspecified, or requires user input before any action can be taken. Do not plan or implement anything—only ask questions to collect the information needed.
Contradiction analysis: Break complex problems into opposing forces, identify the principal contradiction that determines the others and its dominant side, classify it as antagonistic or non-antagonistic, and choose an approach accordingly. Trigger when a problem is tangled, factors constrain one another, priorities are unclear, the root cause is unknown, fixes repeatedly fail, or trade-offs are hard to explain. Do not trigger for direct execution tasks or when the user has already chosen a solution. English: Contradiction analysis. Decompose a tangled problem into opposing forces, isolate the principal contradiction and its dominant side, classify it as antagonistic or non-antagonistic, and choose the response accordingly. Trigger on unclear priorities, unknown root causes, recurring failures, or trade-offs that resist explanation; skip for direct execution or when the user has already chosen the solution.
Set up or update the agent-first engineering harness for any repository. Implements the complete scaffolding that makes AI coding agents effective: knowledge maps (AGENTS.md as a concise TOC), structured documentation, architecture boundaries, enforcement rules (.harness/*.yml specs), quality scoring, and process patterns for agent-driven development. Use this skill whenever someone wants to make a repo agent-ready, set up AGENTS.md or docs/ structure, define domain boundaries or golden principles, generate .harness/ configuration, audit agent readiness, or update an existing harness. Also trigger when a user reports problems with agent effectiveness, context management, or architectural drift — these are symptoms of a missing or stale harness. Trigger on: "harness this repo", "set up harness", "agent-first setup", "make this agent-ready", "update the harness", "assess agent readiness", "set up AGENTS.md", "organize for agents", or any discussion about structuring a codebase for AI agent workflows.
Set up JARVIS Mission Control v2.1.0 — a free, open-source AI agent coordination hub by MissionDeck.ai. Kanban board, real-time WebSocket updates, team chat, scheduled job visibility, agent SOUL editor, Claude Code session tracking, GitHub Issues sync, SQLite-backed webhook delivery with circuit breaker, CSRF + rate limiting. Fork the repo, start the server, open the dashboard. No cloud account required. Instruction-only skill — all code runs on YOUR server from open-source GitHub repo. No code execution in skill. Security-audited: 0 HIGH, 0 CRITICAL. See SECURITY.md for full audit.
Build ChatGPT apps with MCP servers on Cloudflare Workers. Extend ChatGPT with custom tools and interactive widgets (HTML/JS UI). Use when: developing ChatGPT extensions, implementing MCP servers, or troubleshooting CORS, widget 404s, MIME types, ASSETS binding errors, Next.js integration issues, or edge platform limitations.
Use when addressing open PR review comments from any reviewer (human or bot). Runs an in-session loop by default; on an explicit `--handoff` request, completes one unit of work per session and stops so the user starts the next. When an external shell loop should drive the sessions instead of the user, use ralph-wiggum-loop.
Optional command line and MCP server for a macOS app, so Claude Code, Codex, and scripts can drive it (list, add, change, search, act) through the running app. Use when the user wants agents to work with their app or its data, asks for a CLI, an MCP server, or automation for the app, or says "Claude should be able to use my app".
Use when asked to watch, babysit, audit, review, compare, or fix another agent's work from a Codex session ID, Claude Code session/transcript, chat/thread link, PR, branch, log, or pasted run summary. Monitor until the other agent is done or blocked, reconstruct what the user asked, independently investigate the same problem to form your own hypotheses and approach, inspect what the agent actually changed and verified, compare the two investigations, report gaps and add-on directions, and optionally make scoped fixes when the user authorizes repair.
Configure the Canvas Authoring MCP server for the current coauthoring session. Use before starting to edit a canvas app with the /canvas-app skill if not already connected. Use when users say "configure MCP", "set up MCP server", "MCP not working", "connect Canvas Apps MCP", "canvas-authoring not available", "MCP not configured", or "set up canvas apps".
Use when the user asks to “create a hook,” “add a PreToolUse/PostToolUse/Stop hook,” “validate tool use,” “implement prompt-based hooks,” “use ${CLAUDE_PLUGIN_ROOT},” “set up event-driven automation,” “block dangerous commands,” or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks, focusing on the advanced prompt-based hooks API
Autonomous loop that works through docs/STORIES.md one story at a time for a macOS app, building, snapshotting, comparing against reference screenshots, fixing, and committing, until a milestone is done. Use when the user says "keep going", "work through the backlog", "make it match the reference", or wants an unattended improvement loop.
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
Multi-agent research orchestration: split a research goal into parallel sub-goals, run each via headless `claude -p` subprocesses, and aggregate results into a polished report file. Use for systematic web/document research, competitive or industry analysis, batch link/dataset processing, and long-form evidence synthesis. Triggers: "深度调研", "deep research", "wide research", "多 Agent 调研", "系统调研".
This skill should be used for persistent semantic memory in agent systems: cross-session knowledge retention, entity tracking, temporal validity, graph or vector retrieval, memory consolidation, and memory benchmark selection. Route file-backed scratchpads to filesystem-context, handoff summaries to context-compression, and token-efficiency tactics to context-optimization
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
Scaffolds a complete agent TUI in TypeScript using @openrouter/agent — like create-react-app for terminal agents. Generates a customizable terminal interface with three input styles, four tool display modes, ASCII banners, streaming output, session persistence, and configurable tools. Use when building an agent, creating a TUI, scaffolding an agent project, or building a coding assistant.
Discover (by searching) to use thousands of AI agents, MCP, tools, data sources, and APIs on the use-agently.com AI Agent Marketplace/Directory. You can get more effective "Web Search", "Web Fetch" or "Social Search" with use-agently. Sending messages via the A2A protocol, calling MCP tools, making HTTP/Web requests, or interacting with paid agents supporting x402 micropayments.
Use when you need to generate a checklist document with embedded agents inventory, following the embedded template exactly and producing INVENTORY-AGENTS-JAVA.md in the project root. This should trigger for requests such as Create embedded agents inventory checklist; Generate INVENTORY-AGENTS-JAVA.md; Use @002-agents-inventory; Inventory embedded Java agent definitions; List generated agent roles for Java development. Part of Plinth Toolkit
Run the Ouroboros specification-first development loop: reduce ambiguity with a Socratic interview grounded in live git data (commits, churn, contributors), freeze an immutable seed/spec, render the execution plan through spec-kit (/speckit.plan → /speckit.tasks), execute against that contract through cli-anything agent-native CLI harnesses (cli-hub, --json evidence), verify before claiming success, and keep looping until completion is actually verified. Use when the user wants spec-first clarification, git-aware interviews, immutable requirements, drift-aware implementation, harness-driven execution, or a persistent completion loop that should keep going until tests / checks / acceptance criteria pass. Triggers on: ooo, ouroboros, interview, seed, run workflow, evaluate, evolve, ooo ralph, specification first, socratic interview, git-aware interview, ambiguity reduction, execution plan, cli harness execute, persistent completion.
Captures and maintains learnings, errors, and corrections. Use when a command fails, the user corrects an assumption, a capability is missing, knowledge is outdated, or a better approach is found. Review relevant saved guidance before major tasks, reuse, and promotion; revalidate it according to what could change and how quickly.
Deep strategic thinking mode that finds the single highest-leverage, most innovative action by blending concepts across domains. Use this skill whenever the user asks you to think, brainstorm, strategize, or figure out what to do next — even casually. Trigger on phrases like "what should we do", "what's the best approach", "what would you suggest", "think about this", "what's the smartest move", "I'm stuck", "ideas?", "hmm what if we...", "what's next", "how should we approach", or any request for creative/strategic ideation rather than straightforward execution. When in doubt about whether the user wants execution or ideation, lean toward triggering this skill
Use the MemOS Local memory system to search and use the user's past conversations. Use this skill whenever the user refers to past chats, their own preferences or history, or when you need to answer from prior context. When auto-recall returns nothing (long or unclear user query), generate your own short search query and call memory_search. Available tools: memory_search, memory_get, memory_write_public, memory_share, memory_unshare, task_summary, skill_get, skill_search, skill_install, skill_publish, skill_unpublish, network_memory_detail, network_skill_pull, network_team_info, memory_timeline, memory_viewer.
Adopt `oh-my-agent` as a cross-vendor portable multi-agent harness. Use when the user wants a single `.agents/` source of truth that projects into any AI IDE or runtime — Claude Code, Codex, Gemini CLI, Antigravity, Cursor, Grok Build, Kimi Code, OpenCode, Pi, Qwen Code, Reasonix, Amp, GitHub Copilot, or Kiro CLI — without duplicating agent definitions. Also use when mapping Claude workflows (`team`, `autopilot`, `ultrawork`, `ultraqa`) into OMA equivalents for any vendor. Triggers on: ohmg, oh-my-agent, oma, portable harness, .agents, cross-vendor agents, Gemini multi-agent, Antigravity agent team, autopilot, ultrawork, ultraqa, claude code agents, codex agents, cursor agents. Route Claude Code–native runtime orchestration to `omc` and Codex CLI–native runtime orchestration to `omx`.
Create and scaffold plugin directories for Codex with a required `.codex-plugin/plugin.json`, optional plugin folders/files, and baseline placeholders you can edit before publishing or testing. Use when Codex needs to create a new personal plugin, add optional plugin structure, or generate or update personal or repo-root `.agents/plugins/marketplace.json` entries for plugin ordering and availability metadata.
Clarify ambiguous requirements through focused dialogue before implementation. Use when requirements are unclear, features are complex (>2 days), or involve cross-team coordination. Ask two core questions - Why? (YAGNI check) and Simpler? (KISS check) - to ensure clarity before coding.
Run bounded, evidence-driven training research through W&B Launch: assess project readiness, establish launchable code and queue capacity, smoke-test real jobs, execute serial trials, compare metrics, and persist resumable research state. Use when a coding agent is asked to autonomously test training hypotheses or tune a real W&B-tracked workload.
Create, manage, and orchestrate AI agents using the AI Maestro CLI. Use when the user asks to "create agent", "list agents", "delete agent", "hibernate agent", "wake agent", "install plugin", "show agent", "restart agent", or any agent lifecycle management task.
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.