MCP Test MCP

Instructions

Implementation Reference

• node-wrapper/python-src/src/mcp_test_mcp/tools/connection.py:242-328 (handler)

  The core handler implementation for the 'get_connection_status' MCP tool. This async function retrieves the current connection state from ConnectionManager.get_status(), computes metadata like request time and connection duration, handles connected/disconnected cases, provides user-facing ctx updates, logs appropriately, and returns a standardized response dictionary. Error handling ensures graceful degradation even on unexpected exceptions.
  @mcp.tool async def get_connection_status(ctx: Context) -> dict[str, Any]: """Check the current MCP server connection state. Returns detailed information about the active connection including server information, transport type, connection duration, and usage statistics. Returns: Dictionary with connection status including: - success: Always True - connected: Boolean indicating if currently connected - connection: Full ConnectionState if connected, None otherwise - message: Human-readable status message - metadata: Request timing and connection duration info """ start_time = time.perf_counter() try: state = ConnectionManager.get_status() connected = state is not None elapsed_ms = (time.perf_counter() - start_time) * 1000 metadata: dict[str, Any] = { "request_time_ms": round(elapsed_ms, 2), } if connected and state: # Calculate connection duration if state.connected_at: duration_seconds = ( time.perf_counter() - state.connected_at.timestamp() ) metadata["connection_duration_seconds"] = round(duration_seconds, 2) message = f"Connected to {state.server_url}" connection_data = state.model_dump(mode="json") # User-facing debug update await ctx.debug(f"Connection status: connected to {state.server_url}") # Detailed technical log logger.debug( "Connection status checked", extra={ "connected": True, "server_url": state.server_url, "statistics": state.statistics, }, ) else: message = "Not connected to any MCP server" connection_data = None # User-facing debug update await ctx.debug("Connection status: not connected") # Detailed technical log logger.debug("Connection status checked", extra={"connected": False}) return { "success": True, "connected": connected, "connection": connection_data, "message": message, "metadata": metadata, } except Exception as e: # Status check should never fail, but handle gracefully elapsed_ms = (time.perf_counter() - start_time) * 1000 # User-facing error update await ctx.error(f"Unexpected error checking connection status: {str(e)}") # Detailed technical log logger.exception("Unexpected error checking connection status") return { "success": True, # Still return success with disconnected state "connected": False, "connection": None, "message": f"Unable to determine connection status: {str(e)}", "metadata": { "request_time_ms": round(elapsed_ms, 2), "error": str(e), }, }
• node-wrapper/python-src/src/mcp_test_mcp/server.py:96-99 (registration)

  Registration of tools by importing the tools modules (including connection.py containing get_connection_status). The @mcp.tool decorators in the imported modules automatically register the tools with the FastMCP server instance. This import sequence is critical for tool availability.
  # Import tool modules to trigger decorator registration # This MUST happen after mcp instance is imported but before the server runs from .tools import connection, tools, resources, prompts, llm
• node-wrapper/python-src/src/mcp_test_mcp/tools/__init__.py:7-17 (registration)

  Re-export of get_connection_status from connection.py module, making it available when importing from tools package. This facilitates the server.py import of the tools package.
  from .connection import connect_to_server, disconnect, get_connection_status from .tools import call_tool, list_tools __all__ = [ "connect_to_server", "disconnect", "get_connection_status", "list_tools", "call_tool", ]