YouTube Music MCP has 7 tools with complete input schemas and descriptions in Spanish. All tools are READ_ONLY except export_liked_songs (WRITE). Naming follows verb_noun convention (list_, get_, search_, export_). However, descriptions are brief (avg ~80 chars) and lack LLM-optimized guidance on when to use each tool vs. alternatives. Parameter descriptions are present but minimal. Output schemas are not documented, LLMs cannot predict response structure. No error handling guidance, no pagination documentation despite list tools accepting limits. Tool composition is reasonable but lacks chaining hints (e.g., 'get_playlist requires playlist_id from list_library_playlists').

• Document output schemas for all tools. E.g., 'Returns {songs: [{title, artists, album, url, ...}], total: N, limit: M}' so LLMs know what fields to extract.
• Expand descriptions to 100 - 150 chars with WHEN/WHY guidance. E.g., 'list_liked_songs: Retrieve your liked tracks with optional filtering by artist or album. Use this to discover your music taste or export a subset. Requires auth_status to pass first.'
• Add pagination hints: 'Results are limited to the specified limit parameter. For large libraries, call multiple times with offset or cursor-based pagination if supported.'
• Include tool chaining hints in descriptions. E.g., 'list_library_playlists returns playlist_id values that can be passed to get_playlist to fetch full track lists.'
• Add error handling guidance: 'If auth_status returns {authenticated: false}, call setup-auth.sh before retrying. If export_liked_songs fails with permission error, check file system write access to Documents/YouTube Music Exports.'