OpenClaw Skills
extend your agent's capabilities
Skills are pre-packaged capabilities for OpenClaw agents. Each skill comes with instructions, scripts, and configuration that teach your AI how to use specific tools.
💻 Development
coding-agentRun Codex CLI, Claude Code, or other coding agents via background process for programmatic control.
githubInteract with GitHub using the gh CLI. Issues, PRs, CI runs, and API queries.
⚡ Productivity
summarizeSummarize or extract text/transcripts from URLs, podcasts, and local files.
weatherGet current weather and forecasts. No API key required.
peekabooCapture and automate macOS UI with the Peekaboo CLI.
🎬 Media
openai-image-genBatch-generate images via OpenAI Images API with a gallery viewer.
openai-whisper-apiTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
video-framesExtract frames or short clips from videos using ffmpeg.
🔌 Integrations
mcporterUse mcporter CLI to list, configure, auth, and call MCP servers/tools directly.
sonoscliControl Sonos speakers — discover, status, play, volume, group.
🛡️ System
healthcheckHost security hardening and risk-tolerance configuration for OpenClaw deployments.
skill-creatorCreate or update AgentSkills. Design, structure, and package skills with scripts and assets.
Installing Skills
Build Your Own Skills
The AI Dev Team playbook includes the skill-creator workflow and templates for building custom skills.
Get the Playbook — $29