Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Cross-platform operating system automation and screen control toolkit. Use when users need screenshots, mouse/keyboard control, visual recognition, window management, browser automation, or desktop automation tasks. Supports macOS 12+ and Windows 10+. On macOS, uses AppleScript, pyautogui, and OpenCV. On Windows, uses pywinauto, pyautogui, and OpenCV (no Hammerspoon equivalent).
Spark a prairie fire: When starting from almost nothing, assess the conditions and choose one base area—a minimal foothold that can hold, be reused, and expand. Dig in, then grow step by step instead of scattering effort. Trigger for new projects, MVPs, pilots, daunting goals with no obvious first step, or busy work that accumulates nothing; skip when a solid base already exists or the task modifies an existing system.
Workflow patterns: When a task clearly needs several skills in sequence, pick one of four standard patterns—project kickoff, hard-problem assault, iterative improvement, or stage retrospective—that define the order, handoff artifact for each step, and stop condition. Trigger for starting from scratch, tackling a difficult problem, iterative improvement, or a stage retrospective; skip when a single skill suffices.
Build autonomous game-playing agents using AI and reinforcement learning. Covers game environments, agent decision-making, strategy development, and performance optimization. Use when creating game-playing bots, testing game AI, strategic decision-making systems, or game theory applications.
Design and operate an advanced AI agent memory system with the published HelixDB v3 SDKs using hybrid graph plus exact traversal-prefiltered vector and BM25 search. Use for long-term memory, profiles, RAG, recall, extraction, deduplication, consolidation, versioning, forgetting, categorisation, or ingestion. TypeScript-first (`@helix-db/[email protected]`); a Rust v3 variant is in EXAMPLES.rust.md.
This skill should be used when working with memory-lancedb-pro, a production-grade long-term memory MCP plugin for OpenClaw AI agents. Use when installing, configuring, or using any feature of memory-lancedb-pro including Smart Extraction, hybrid retrieval, memory lifecycle management, multi-scope isolation, self-improvement governance, or any MCP memory tools (memory_recall, memory_store, memory_forget, memory_update, memory_stats, memory_list, self_improvement_log, self_improvement_extract_skill, self_improvement_review)
Use the `sesh` CLI/TUI to list, find, enter, resume, create, tag, archive, rename, reparent, capture, send-to, delegate, or inspect coding-agent threads across machines. Use when the user asks about sesh command usage, the thread TUI, entering or resuming a thread, cross-machine thread state, the sesh daemon, peers, mycockpit (the cross-machine tmux cockpit), or tickets
Meta-skill for understanding and customizing Mindfold Trellis — the all-in-one AI workflow system for 11 AI coding platforms (Claude Code, Cursor, OpenCode, iFlow, Codex, Kilo, Kiro, Gemini CLI, Antigravity, Qoder, CodeBuddy). Documents the original Trellis system design including architecture, commands, hooks, multi-agent pipelines, monorepo support, and task lifecycle hooks. Use when understanding Trellis architecture, customizing workflows, adding commands or agents, troubleshooting issues, or adapting Trellis to specific projects. Modifications should be recorded in a project-local trellis-local skill, not here.
Evaluate and adopt Unity game-development skill packs from external repositories into a safe, reusable local package. Use when maintainers want to import game-dev workflows (Addressables, Cinemachine, GAS, VContainer, UniTask, Wwise, etc.) without blindly trusting third-party prompts.
Universal fallback for executing actions across 1,000+ apps when no other skill is available. Use this skill ONLY when the user requests an action on an app/service not covered by other skills (e.g., Notion, Asana, Trello, HubSpot, Airtable, Linear, Monday, Zendesk, Intercom, Stripe, Shopify, QuickBooks, Zoom, Microsoft 365, Dropbox, Box, Figma, Jira, Confluence, etc.). Do NOT use if another skill already handles the service. Triggers on requests to connect to external apps, execute actions on third-party services, or when user asks "can you actually do X" for an unsupported service.
BMad Autonomous Development — orchestrates parallel story implementation pipelines. Builds a dependency graph, updates PR status from GitHub, picks stories from the backlog, and runs each through create → dev → review → PR in parallel — each story isolated in its own git worktree — using dedicated subagents with fresh context windows. Loops through the entire sprint plan in batches, with optional epic retrospective. Use when the user says "run BAD", "start autonomous development", "automate the sprint", "run the pipeline", "kick off the sprint", or "start the dev pipeline". Run /bad setup or /bad configure to install and configure the module.
Gets Claude talking to a tool that has no official connector. Identifies the system, first checks Claude's connector directory for an existing connector, and if none exists, connects the tool through Zapier—never by hand-building against a raw API—with credentials handled safely and approval gates for anything that writes. Turns “my ERP isn't supported” into an afternoon. Use whenever the owner names a tool that isn't connected, including “can you connect to my,” “make my tools talk to each other,” “there's no API for this,” “pull data out of my field service software,” “I use ServiceTrade and it's not on the list,” or “is there any way to get at this data.” Use whenever an unsupported system is blocking something else.
Monitors context window health during large, long-running, multi-session, or explicitly context-sensitive work. Uses bounded external anchors, one cold-context review when self-recovery is uncertain, and a clean handoff on real drift. Do not auto-activate merely because a medium task has an intent frame and plan.
The meta skill. Turns any raw feature into a properly skilled, tested, resolvable unit of agent capability. Idempotent: running it on an existing skill improves it (bug fix, new input, quality pass) instead of starting from scratch. Every skill declares an EVAL CONTRACT (its goal + skill-specific dimensions + hard-fails) so the cross-modal eval judges THIS skill's real purpose, not generic slop. Cross-modal eval runs BEFORE tests: 3 frontier models from different providers critique the output against the contract; you iterate to quality, THEN write or update tests that lock in the proven-good behavior. NO-REGRESSION LAW: any edit to a skill must score >= the previous iteration's eval — forward only, never back. For skills that back a scheduled job, an edit MUST re-run a representative task and eval it before shipping.
Create agent company packages conforming to the Agent Companies specification (agentcompanies/v1). Use when a user wants to create a new agent company from scratch, build a company around an existing git repo or skills collection, or scaffold a team/department of agents. Triggers on: "create a company", "make me a company", "build a company from this repo", "set up an agent company", "create a team of agents", "hire some agents", or when given a repo URL and asked to turn it into a company. Do NOT use for importing an existing company package (use the CLI import command instead) or for modifying a company that is already running in Paperclip
Use when the user wants to talk through work with Pipa or their current agent in a live voice huddle: Pipa Huddle, Pipa voice session, talk this through, walking work session, plan this by voice, voice call with my agent, or discuss this out loud. Starts a voice huddle with agent context and optional handoff. Do not use for audio brief generation, generic TTS, video meetings, Zoom/Meet bots, or realtime voice coding/permission approval.
Ultra-compressed communication mode. Cuts token usage by about 75% by speaking like a caveman while preserving full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when the user says “caveman mode,” “talk like caveman,” “use caveman,” “less tokens,” “be brief,” or invokes /caveman. Also auto-triggers when token efficiency is requested.
Build and deploy AI agents with CloudBase Agent SDK (TypeScript & Python). Implements the AG-UI protocol for streaming agent-UI communication. Use when deploying agent servers, using LangGraph/LangChain/CrewAI adapters, building custom adapters, understanding AG-UI protocol events, or building web/mini-program UI clients. Supports both TypeScript (@cloudbase/agent-server) and Python (cloudbase-agent-server via FastAPI).
Show real token usage and estimated savings for the current session. Reads directly from the Claude Code session log — no AI estimation. Triggers on /caveman-stats. Output is injected by the mode-tracker hook; the model itself does not compute the numbers.
Expert skill for OpenClaw (v2026.2.19) — self-hosted AI gateway connecting chat apps (WhatsApp, Telegram, Discord, Slack, iMessage, Signal, LINE, Matrix, Teams, Google Chat) to AI agents. Use when user asks about: installing/configuring/updating OpenClaw, setting up or troubleshooting chat channels, security hardening and auditing, inspecting openclaw.json config or prompts (SOUL.md/AGENTS.md), multi-agent routing and session management, cloud deployment (AWS/GCP/Fly.io/Docker), upgrading versions, or discovering ClawHub skills. Includes bundled scripts for security audit, config inspection, and session scanning. Triggers on: "openclaw", "openclaw.json", "~/.openclaw", gateway config. Not for: general chatbot frameworks, non-OpenClaw gateways, or unrelated Docker questions.
Context is the complete state available to a language model at inference time. It includes everything the model can attend to when generating responses: system instructions, tool definitions, retrieved documents, message history, and tool outputs.
Expert guidance for creating, building, and using Claude Code subagents and the Task tool. Use when working with subagents, setting up agent configurations, understanding how agents work, or using the Task tool to launch specialized agents
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Expert knowledge for Microsoft Foundry (aka Azure AI Foundry) development, including troubleshooting, best practices, decision-making, architecture and design patterns, limits and quotas, security, configuration, integrations and coding patterns, and deployment. Use when building Foundry agents with Azure OpenAI, voice/telephony, Realtime, MCP tools, or LangChain integrations, and for other Microsoft Foundry-related development tasks. Not for Content Safety in Foundry Control Plane (use azure-content-safety), Azure Content Understanding in Foundry Tools (use azure-content-understanding), Azure Speech in Foundry Tools (use azure-speech), or Microsoft Foundry Classic (use microsoft-foundry-classic)
Anti-detection browser automation for AI agents. Use when the user needs stealth web browsing, undetectable scraping, fingerprint spoofing, proxy rotation, or privacy-focused browser automation. Triggers include "stealth scrape", "anti-detection", "bypass fingerprinting", "camofox", "camoufox", "undetectable browser", "bot evasion", or any browser task requiring evasion of bot detection systems.
Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress FILEPATH or "compress memory file"
Replace with a trigger-style description of when this skill should activate. Be specific — this is what the agent uses to decide whether to load the skill. Example: "Sui TypeScript SDK integration. Use when writing, reviewing, or debugging TypeScript code that interacts with Sui RPCs, transactions, or on-chain state."
Decision guide for delegating to caveman-style subagents. Tells the main thread WHEN to spawn cavecrew-investigator (locate code), cavecrew-builder (1-2 file edit), or cavecrew-reviewer (diff review) instead of doing the work inline or using vanilla Explore. Subagent output is caveman-compressed so the tool-result injected back into main context is ~60% smaller — main context lasts longer across long sessions. Trigger: "delegate to subagent", "use cavecrew", "spawn investigator/builder/reviewer", "save context", "compressed agent output"
Scaffolds a headless agent in TypeScript using @openrouter/agent and Bun — for CLI tools, API servers, queue workers, and pipelines. No terminal UI. Use when building a headless agent, programmatic agent, CLI tool that uses AI, batch agent, pipeline agent, API agent, agent without a UI, or agent service.
MANDATORY pre-flight check before answering any question or writing any code related to Meta Quest VR headsets (Quest 2, Quest 3, Quest 3S, Quest Pro) or apps that target them. Forces verification against authoritative Meta sources via the metavr CLI / metavr MCP tools BEFORE relying on training-data knowledge. Counters the failure mode where agents answer Quest-related questions from stale memory and ship deprecated APIs, broken Android manifests, and store-rejected builds. Loads automatically when the user is in a Quest project (any reference to Oculus / Meta Quest / Horizon OS, Unity OVR or Meta XR packages, com.meta.* / com.oculus.* package IDs, Quest-targeted AndroidManifest, .meta files, or developers.meta.com / developer.oculus.com URLs). Build paths: all; use this skill first whenever the path is unknown or a Quest-specific claim is involved.
Design, test, create, and attach LangSmith online evaluators for production traces or conversation threads. Use for workspace evaluators, run or thread rules, sampling, filters, backfills, and evaluator monitoring; use eval-engineering for Harbor tasks and agent benchmarks.
Re-run or deepen the agent's identity interview. Drives the SHARED bootstrap answer bank (state/interview.json) via `gbrain bootstrap interview --set/--show/--confirm`, then re-renders the identity files with `gbrain bootstrap render --only <FILE> --force`. One answer store, two surfaces: first-boot bootstrap and this re-run skill write the same bank, so an updated answer flows into SOUL.md, USER.md, ACCESS_POLICY.md, and HEARTBEAT.md without hand-editing rendered files. Every answer is the user's own words — never invented.
Use this skill when auditing a codebase for AI agent readiness, or when guiding improvements to make a codebase work well with agentic coding tools. It applies when users ask to evaluate test coverage, file structure, type system usage, dev environment speed, or automated enforcement—the five pillars that determine how effectively coding agents can operate in a project. Triggers on “audit my codebase,” “make this agent-ready,” “improve for AI agents,” “agent-friendly,” or questions about why agents struggle with a codebase.
Fill-in-the-blank prompt templates for delegating work to subagents or tmux workers — SEARCH/LOCATE, IMPLEMENT, REFACTOR, RESEARCH, REVIEW/VERIFY. Invoke BEFORE writing any delegation prompt (Agent tool, agent-tmux worker, or fanout task) so the prompt ships with explicit GOAL, ACCEPTANCE, and REPORT sections instead of a vague ask. Not for deciding WHETHER to delegate or for driving workers after launch (see the using-tmux-agent-tools and tmux-agent-tools skills for those).
Set up Symphony (OpenAI's Codex orchestrator) for a user's repo. Use when the user mentions Symphony setup, configuring Symphony, getting Symphony running, or wants to connect their repo to Linear for autonomous Codex agents. Also use when the user says "set up symphony", "configure symphony for my repo", or references WORKFLOW.md configuration.
Answer questions about the AI SDK and help build AI-powered features. Use when developers: (1) Ask about AI SDK functions like generateText, streamText, ToolLoopAgent, embed, or tools, (2) Want to build AI agents, chatbots, RAG systems, or text generation features, (3) Have questions about AI providers (OpenAI, Anthropic, Google, etc.), streaming, tool calling, structured output, or embeddings, (4) Use React hooks like useChat or useCompletion. Triggers on: "AI SDK", "Vercel AI SDK", "generateText", "streamText", "add AI to my app", "build an agent", "tool calling", "structured output", "useChat".
This skill guides building AI agent endpoints on EdgeOne Makers — five framework routes (DeepAgents, LangGraph, CrewAI, OpenAI Agents SDK, Claude Agent SDK), platform-injected `context.store` / `context.tools` / `context.sandbox`, conversation_id dual-channel routing, SSE streaming, and `agents/` vs `cloud-functions/` separation. It should be used when the user wants to create or review an AI agent endpoint on EdgeOne Makers — e.g. "build an agent on EdgeOne Makers", "create a Claude agent endpoint", "wire LangGraph into Makers", "stream LLM responses with SSE", "review my agent template", "use context.store / context.sandbox / context.tools". Do NOT trigger for plain Edge Functions, Cloud Functions, or middleware (those don't run AI logic — use edgeone-pages-dev instead). Do NOT trigger for deployment workflows (use edgeone-pages-deploy). Do NOT trigger for generic AI framework development outside an EdgeOne Makers project.
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.