ElevenLabs MCP Server

Instructions

⚠️ COST WARNING: This tool makes an API call to ElevenLabs which may incur costs. Only use when explicitly requested by the user.

Implementation Reference

• elevenlabs_mcp/server.py:469-474 (registration)

  The @mcp.tool decorator registers the 'isolate_audio' tool, including its description which implies the input schema (input_file_path: str, output_directory: str | None = None).

      description="""Isolate audio from a file and save the output audio file to a given directory.
      Directory is optional, if not provided, the output file will be saved to $HOME/Desktop.
  
      ⚠️ COST WARNING: This tool makes an API call to ElevenLabs which may incur costs. Only use when explicitly requested by the user.
      """
  )

• elevenlabs_mcp/server.py:475-494 (handler)

  The core implementation of the 'isolate_audio' tool. It processes the input audio file, calls the ElevenLabs client.audio_isolation.convert API to isolate the audio, saves the result to an output file, and returns a success message with the file path.

  def isolate_audio(
      input_file_path: str, output_directory: str | None = None
  ) -> TextContent:
      file_path = handle_input_file(input_file_path)
      output_path = make_output_path(output_directory, base_path)
      output_file_name = make_output_file("iso", file_path.name, output_path, "mp3")
      with file_path.open("rb") as f:
          audio_bytes = f.read()
      audio_data = client.audio_isolation.convert(
          audio=audio_bytes,
      )
      audio_bytes = b"".join(audio_data)
  
      with open(output_path / output_file_name, "wb") as f:
          f.write(audio_bytes)
  
      return TextContent(
          type="text",
          text=f"Success. File saved as: {output_path / output_file_name}",
      )