get_track_radio
Discover similar tracks to any TIDAL song to expand playlists and find new music based on your current listening preferences.
Instructions
Get tracks similar to a seed track (track radio).
This returns TIDAL's native recommendations based on the specified track, useful for music discovery and creating "similar music" playlists.
Args: track_id: ID of the seed track limit: Maximum tracks to return (default: 20, max: 100)
Returns: List of similar tracks with seed track info
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| track_id | Yes | ||
| limit | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| count | Yes | Number of tracks returned | |
| status | Yes | Operation status (success/error) | |
| tracks | Yes | List of recommended tracks | |
| seed_id | Yes | ID of the seed track or artist | |
| seed_name | Yes | Name of the seed track or artist | |
| seed_type | Yes | Type of seed (track/artist) |
Implementation Reference
- src/tidal_mcp/server.py:829-886 (handler)The handler function decorated with @mcp.tool() that implements the get_track_radio tool. It authenticates, fetches the seed track, calls tidalapi's get_track_radio, maps to Track models, and returns RadioTracks.
@mcp.tool() async def get_track_radio(track_id: str, limit: int = 20) -> RadioTracks: """ Get tracks similar to a seed track (track radio). This returns TIDAL's native recommendations based on the specified track, useful for music discovery and creating "similar music" playlists. Args: track_id: ID of the seed track limit: Maximum tracks to return (default: 20, max: 100) Returns: List of similar tracks with seed track info """ if not await ensure_authenticated(): raise ToolError("Not authenticated. Please run the 'login' tool first.") try: limit = min(max(1, limit), 100) # Get the seed track first track = await anyio.to_thread.run_sync(session.track, track_id) if not track: raise ToolError(f"Track with ID '{track_id}' not found") seed_name = f"{track.name} by {track.artist.name if track.artist else 'Unknown Artist'}" # Get radio tracks radio_tracks = await anyio.to_thread.run_sync( lambda: track.get_track_radio(limit=limit) ) tracks = [] for t in radio_tracks: tracks.append( Track( id=str(t.id), title=t.name, artist=t.artist.name if t.artist else "Unknown Artist", album=t.album.name if t.album else "Unknown Album", duration_seconds=t.duration, url=f"https://tidal.com/browse/track/{t.id}", ) ) return RadioTracks( status="success", seed_id=track_id, seed_type="track", seed_name=seed_name, count=len(tracks), tracks=tracks, ) except ToolError: raise except Exception as e: raise ToolError(f"Failed to get track radio: {str(e)}") - src/tidal_mcp/models.py:209-218 (schema)Pydantic BaseModel defining the structured output schema for the get_track_radio tool response, including seed info and list of recommended tracks.
class RadioTracks(BaseModel): """Radio/recommendation tracks based on a seed track or artist.""" status: str = Field(description="Operation status (success/error)") seed_id: str = Field(description="ID of the seed track or artist") seed_type: str = Field(description="Type of seed (track/artist)") seed_name: str = Field(description="Name of the seed track or artist") count: int = Field(description="Number of tracks returned") tracks: List[Track] = Field(description="List of recommended tracks") - src/tidal_mcp/server.py:829-829 (registration)The @mcp.tool() decorator registers the get_track_radio function as an MCP tool.
@mcp.tool() - src/tidal_mcp/server.py:858-860 (helper)Helper lambda that calls the underlying tidalapi Track.get_track_radio method to fetch similar tracks.
radio_tracks = await anyio.to_thread.run_sync( lambda: track.get_track_radio(limit=limit) )