Skip to content

MCP API Reference

This section provides details about the Model Context Protocol (MCP) API endpoints implemented by Adaptive Graph of Thoughts.

Overview

Adaptive Graph of Thoughts implements an MCP server to allow integration with MCP clients such as Claude Desktop. The primary endpoint for MCP communication is /mcp.

Methods

The following JSON-RPC methods are supported via the /mcp endpoint:

initialize

  • Description: Initializes the session between the MCP client and the Adaptive Graph of Thoughts server. Provides server information to the client.
  • Parameters: See MCPInitializeParams in src/asr_got_reimagined/api/schemas.py.
  • Returns: MCPInitializeResult (see src/asr_got_reimagined/api/schemas.py).

asr_got.query

  • Description: Submits a natural language query to the Adaptive Graph of Thoughts ASR-GoT engine for processing through its 8-stage pipeline.
  • Parameters: See MCPASRGoTQueryParams in src/asr_got_reimagined/api/schemas.py. This includes:
    • query (str): The main query string.
    • session_id (Optional[str]): An optional session identifier.
    • parameters (Optional[ASRGoTQueryExecutionParams]): Parameters to control query execution, such as:
      • include_reasoning_trace (bool): Whether to include a summary of stage outputs.
      • include_graph_state (bool): Whether to include the full graph state in the response.
      • Other operational parameters specific to ASR-GoT stages.
    • context (Optional[ASRGoTQueryContext]): Optional initial context for the query.
  • Returns: MCPASRGoTQueryResult (see src/asr_got_reimagined/api/schemas.py). This includes:
    • answer (str): The final answer generated by the pipeline.
    • reasoning_trace_summary (Optional[str]): A summary of the reasoning steps if requested.
    • graph_state_full (Optional[GraphStateSchema]): The full graph state if requested.
    • confidence_vector (Optional[list[float]]): The final confidence assessment.
    • session_id (str): The session ID for this query-response cycle.
    • execution_time_ms (int): The total processing time for the query.

shutdown

  • Description: Signals the server that the client session is ending and the server can prepare for shutdown or resource cleanup if necessary.
  • Parameters: See ShutdownParams in src/asr_got_reimagined/api/schemas.py.
  • Returns: None.

Further details on request/response schemas can be found by examining the Pydantic models in src/asr_got_reimagined/api/schemas.py.