What is music-discover?

Music Discover is a personalized AI assistant that integrates with Spotify to deliver tailored music news, new release alerts, and smart recommendations. By analyzing your listening history and top artists, it automates the discovery of new tracks and keeps you informed about upcoming concerts and artist updates, providing a seamless bridge between your music library and real-time industry news.

When should I use music-discover?

music-discover is useful in the following scenarios: • Stay Informed: Automatically track new album releases, singles, and major concert announcements for your most-listened-to artists. • Contextual Recommendations: Generate custom music suggestions tailored to specific activities or moods, such as commuting, working out, or traveling. • Effortless Library Management: Create 'Discoveries' playlists or add recommended tracks directly to your Spotify queue through simple chat commands. • Taste Exploration: Discover 'deep cuts' from favorite artists and explore new genres based on your unique listening profile and medium-term habits.

name	music-discover
description	Music news and recommendations based on Spotify listening history. Use for music news, new releases, recommendations, what to listen to in specific situations (commute, workout, travel), artist updates, or concert announcements.

Music Discover

Personalized music news and recommendations using Spotify data and listening history.

TOOL RESTRICTION: Never use the webFetch tool for web searching. Use the model's built-in web search if available.

Workflow

Step 1: Fetch User's Music Profile

Use spotifyUser tool:

get_top_tracks (timeRange: short_term) → recent favorites
get_top_artists (timeRange: medium_term) → overall taste

Step 2: Find New Releases

Use spotifyArtists tool on top 2-3 artists:

get_artist_albums (includeGroups: ["album", "single"], limit: 5) → recent releases from favorite artists

Step 3: Enhance with Search

Use spotifySearch tool for specific queries:

search (query: "[users mood, genre or interests]", types: ["track", "artist"]) → find music matching specific moods or genres

Step 4: Search for News

If web search is available: Search for recent news about the user's top artists:

New album releases (last 4 weeks)
Upcoming album announcements
Concert tour announcements (especially Berlin/Germany)
Major artist news (collaborations, announcements)

Skip the news section if no relevant news found.

If web search is NOT available:

Tell the user: "Web search is not enabled. Enable it in settings for current music news."
Continue without news section.

Step 5: Generate Recommendations

Provide 3-5 recommendations based on:

New releases from favorite artists (from Step 2)
Deep cuts from known artists (lesser-known albums)
Genre exploration based on top artists' genres
Search results using spotifySearch tool for specific queries or mood-based searches

Tailor to context if specified (e.g., "long train ride" → longer, atmospheric albums).

Step 6: Offer Actions (Optional)

After showing recommendations, ask if user wants to:

Play now or add to queue → use spotifyPlayback (play) or spotifyQueue (add_to_queue)
Create a "Discoveries" playlist → use spotifyPlaylists (create_playlist + add_tracks)

Output Format

[Brief excited intro with emoji]

News: (only if web search available and news found)
- Artist - "Title/Event" (Date) - Brief description [Link]
...

Recommendations:
- Artist - "Album/Song" - Why this fits [Link if available]
...

[Optional: Offer to play, queue, or create playlist]

Keep it short: Readable in under 30 seconds.

Error handling

If you are trying to use a spotify tool and it fails, guide the user to check their settings:

Tell user to check if Spotify is connected. The user can connect it from the sidebar menu
Tell user to check if spotify tools for discovery and research are enabled. The user can enable them in the chat input options
Tell the user to try again after they have done the above

music-discover

When & Why to Use This Skill

Use Cases