Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
AI-powered music research with 92+ tools across 17 sources — MusicBrainz, Bandcamp, Discogs, Genius, Last.fm, Wikipedia, and more. Influence tracing, track v...
AI-powered music research with 92+ tools across 17 sources — MusicBrainz, Bandcamp, Discogs, Genius, Last.fm, Wikipedia, and more. Influence tracing, track v...
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
You have access to Crate's music research tools via MCP. These tools connect to 17 real music databases and 26 publications. Use them to answer music questions with verified, cited data.
Add Crate as an MCP server in your configuration: { "mcpServers": { "crate": { "command": "npx", "args": ["-y", "crate-cli", "--mcp-server"], "env": { "ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}" } } } } This exposes all active tools over stdio. Additional API keys unlock more servers (see Optional API Keys below).
Cross-reference multiple sources for comprehensive artist profiles: musicbrainz_search_artist — canonical artist ID, discography, relationships genius_get_artist — bio, aliases, social links, annotations lastfm_get_artist_info — listening stats, similar artists, tags discogs_search_artist — label history, pressings, catalog numbers bandcamp_search — independent releases, merch, direct-support links wikipedia_search — biographical context, career timeline Always start with MusicBrainz for the canonical ID, then fan out to other sources.
Discover how artists connect through published music criticism: Use influence_trace_influence to search 26 publications for co-mentions Results include publication name, critic byline, date, and URL for every connection Use influencecache_get_path for cached paths (instant BFS results) Use influencecache_get_neighbors to explore an artist's immediate connections The influence graph grows with every query — cached in local SQLite Always cite the publication and review when presenting influence connections. Every claim needs a URL.
CRITICAL: Never invent track names. Always verify tracks exist before presenting them. bandcamp_get_artist_tracks — primary source for independent artists musicbrainz_search_recording — primary source for mainstream releases youtube_search — fallback verification source If a track cannot be verified against any real database, do not include it
discogs_get_release — pressing details, labels, catalog numbers, condition notes discogs_get_master_release — all versions/pressings of an album discogs_get_marketplace_stats — current market prices and trends collection_add_record / collection_search — manage the user's personal collection
Research tracks using the sources above — verify every track exists playlist_create — create a new playlist playlist_add_track — add verified tracks with source URLs playlist_export_m3u — export to M3U format for external players Never include a track that hasn't been confirmed against a real database
Share research as public web pages or blog posts: telegraph_create_page — instant shareable page, no account needed telegraph_create_index — create a living index of all published pages tumblr_create_post — post to the user's Tumblr blog with markdown formatting tumblr_tag_post — auto-tag posts with artist names and genres Always include citations and source links in published research
Every claim must be backed by a real data source — never hallucinate facts, tracks, or connections Influence connections require full attribution: publication name, critic, date, and URL Verify tracks against Bandcamp, MusicBrainz, or YouTube before including in any list Cross-reference facts across multiple sources when possible The influence system searches 26 publications including Pitchfork, The Wire, Resident Advisor, Stereogum, The Guardian, NPR, NME, Bandcamp Daily, and more
ServerToolsEnv RequiredDescriptionMusicBrainz6—Artist/release/recording metadataBandcamp7—Independent music, artist tracksWikipedia3—Biographical contextYouTube6—Video search, audio playbackRadiovaries—Internet radio streamingNewsvaries—Music news via RSSCollection5—Local record collection (SQLite)Playlistvaries—Playlist management (SQLite)Influence Cache8—Local influence graph (SQLite)Telegraph5—Anonymous publishingLast.fm7LASTFM_API_KEYScrobbles, similar artistsGenius8GENIUS_ACCESS_TOKENLyrics, annotationsDiscogs9DISCOGS_KEY, DISCOGS_SECRETVinyl catalog, marketplaceWeb Search4TAVILY_API_KEY or EXA_API_KEYPublication searchInfluence3TAVILY_API_KEY or EXA_API_KEYLive influence tracingTumblr5TUMBLR_CONSUMER_KEY, TUMBLR_CONSUMER_SECRETBlog publishingMemory3MEM0_API_KEYPersistent user preferences
Set these environment variables to unlock additional servers: LASTFM_API_KEY — Last.fm listening stats and similar artists GENIUS_ACCESS_TOKEN — Lyrics, annotations, and artist bios DISCOGS_KEY — Vinyl catalog, labels, and marketplace DISCOGS_SECRET — Required with DISCOGS_KEY TAVILY_API_KEY — Web search across 26 music publications EXA_API_KEY — Neural semantic search for influence tracing YOUTUBE_API_KEY — Improved YouTube search results TUMBLR_CONSUMER_KEY — Publish research to your Tumblr blog TUMBLR_CONSUMER_SECRET — Required with TUMBLR_CONSUMER_KEY MEM0_API_KEY — Persistent memory across sessions Only ANTHROPIC_API_KEY is required. All other servers are optional.
Code helpers, APIs, CLIs, browser automation, testing, and developer operations.
Largest current source with strong distribution and engagement signals.