Purpose
Agent-Skills.md is an agent skills marketplace used to help users find the right skills for AI agents. It is positioned as a curated skills directory that indexes skill repositories, allowing users to browse, preview, and download skills in a unified format.
Main Capabilities
- Collects and indexes skill repositories, providing a unified format for browsing, previewing, and downloading.
- Each skill is stored in a separate folder and described by a SKILL.md file, which contains metadata and instructions.
- By default, it scans the skills/ folder in a repository; it also supports submitting URLs that point directly to a specific skill folder.
- Provides popular search examples covering a variety of skill directions, such as: summarizing Codex or Claude CodexBar local cost logs by model; querying Google Places text search, place details, parsing, reviews, or scriptable JSON via goplaces; handling iMessage/SMS direct messages, groups, replies, reactions, polls, listening, and private API operations using the OpenClaw agent's imsg CLI; diagnosing OpenClaw Control UI browser and native Android, iOS, and macOS node connection failures; auditing/hardening OpenClaw hosts (SSH, firewall, updates, exposure surface, backups, disk encryption, gateway security); OpenAI audio transcription API (gpt-4o-transcribe, mini, diarize, or whisper-1); Tavily web search, content extraction, and research tools; initiating voice calls through the OpenClaw voice call plugin; creating, viewing, editing, deleting, searching, moving, or exporting Apple Notes on macOS via the memo CLI; creating, searching, and managing Bear notes via the grizzly CLI; Gemini CLI one-off prompts, summaries, generation, skills, hooks, MCP, or Gemma routing; BluOS CLI (blu) for discovery, playback, grouping, and volume; Foodora dedicated CLI for querying historical orders and active order status; Himalaya CLI for IMAP/SMTP email (list, read, search, compose, reply, forward, copy, move, delete); using nano-pdf CLI edits PDFs with natural language instructions; uses the official obsidian CLI to handle Obsidian vaults (read/search/create/edit notes, tasks, links, properties, plugins); captures frames or clips from RTSP/ONVIF cameras and local webcams, including USB pan/tilt/zoom control; sets up and uses the 1Password CLI for sign-in, desktop integration, and reading or injecting secrets; performs Slack messaging workflows through OpenClaw's messaging tools; monitors blogs and RSS/Atom feed updates with the blogwatcher CLI; sends third-party WhatsApp messages or syncs/searches WhatsApp history via wacli; lists, adds, edits, completes, or deletes Apple Reminders and reminder lists via remindctl; delegates coding work to Codex, Claude Code, or OpenCode as background workers; searches ClawHub skills when required capabilities are unavailable, and installs, verifies, updates, uninstalls, publishes, or syncs skills; controls the Eight Sleep Pod (status, temperature, alarms, schedules); Notion CLI/API for pages, Markdown content, data sources, files, comments, search, Workers, and raw API calls; fetches GitHub issues, selects candidates, generates background fixing agents, opens PRs, and optionally handles PR review comments; lists, configures, authenticates, invokes, and inspects MCP servers/tools over HTTP or stdio via mcporter; captures and automates the macOS UI with the Peekaboo CLI; searches GIF providers, downloads results, and extracts still frames/atlases with a CLI/TUI; controls Sonos speakers (discovery/status/playback/volume/grouping); ElevenLabs text-to-speech with a mac-style say user experience; local speech-to-text with the Whisper CLI (no API key required); xurl CLI for authenticated X posts, replies, reads/searches, DMs, media uploads, followers, auth status, or raw v2 API calls; fetches current weather and forecasts via web_fetch, falling back to wttr.in curl to query locations,Rainfall, temperature, trip planning; play/search Spotify in the terminal via spogo (preferred) or spotify_player; run macOS Parallels smoke tests, including Discord sending, host verification, host replies, and guest read-back proof; write or review AgentSkills: create, fix, validate, or refactor SKILL.md files and bundled resources; check, patch, verify, publish, or confirm OpenClaw GHSA security advisories and private branch status; local text-to-speech via sherpa-onnx (offline, no cloud); review, triage, fix, or land OpenClaw issues and pull requests; Google Workspace CLI for Gmail, Calendar, Drive, Contacts, Sheets, and Docs; summarize or transcribe URLs, YouTube/videos, podcasts, articles, transcripts, PDFs, and local files; manage Trello boards, lists, and cards via the Trello REST API; add, update, list, search, or check Things 3 to-dos, Inbox, Today, projects, areas, and tags on macOS; control tmux sessions/panes for interactive CLI operations (list, capture output, send keys, paste text, monitor prompts); control Philips Hue lights and scenes via the OpenHue CLI; Oracle CLI second-model review/debug/refactor/design, with selected files, dry-run token checks, API or browser engine; investigate OpenClaw pnpm test memory growth, Vitest OOM, RSS spikes, and heap snapshot deltas; prepare, snapshot, run, rerun, debug, or interpret OpenClaw Parallels guest installation, onboarding, gateway smoke tests, and upgrade checks, covering macOS, Windows, and Linux.
- The page shows a total of 1717 pages and 85832 results.
- Agent Skills is supported by leading AI development tools.
How to Use
- Browse, preview, and download skills.
- Submit a repository: click Submit in the header, paste the GitHub URL pointing to the skills folder, and the site will parse and add valid skills to the catalog.
- Skills are in the repositoryStorage location: by default the skills/ folder is scanned, but you can also submit a URL pointing directly to a specific skill folder.
- SKILL.md requirements: must include YAML frontmatter with at least name and description; the body contains the actual guidance and steps for the agent.
Use Cases
Suitable for users who need to find, browse, preview, and download reusable skills for AI agents, as well as developers who want to submit skill repositories to be included.
Limitations
Specific information about fees, pricing, platform support scope, effects, and so on is not yet provided in the materials.