Kiro CLI MCP Server

Instructions

Implementation Reference

• src/kiro_cli_mcp/server.py:372-386 (handler)

  The _handle_task_cancel function that executes the kiro_task_cancel tool logic: extracts task_id from arguments, calls task_manager.cancel_task, and returns success status with message.

  async def _handle_task_cancel(
      task_manager: StreamingTaskManager,
      arguments: dict[str, Any]
  ) -> dict[str, Any]:
      """Handle kiro_task_cancel tool call - cancel a running task."""
      task_id = arguments.get("task_id", "")
      
      cancelled = await task_manager.cancel_task(task_id)
      
      return {
          "success": cancelled,
          "task_id": task_id,
          "message": "Task cancelled" if cancelled else "Task not found or already completed",
      }

• src/kiro_cli_mcp/tools.py:177-190 (schema)

  JSON schema definition for the kiro_task_cancel tool input, requiring a task_id string.

  {
      "name": "kiro_task_cancel",
      "description": "Cancel a running async task",
      "inputSchema": {
          "type": "object",
          "properties": {
              "task_id": {
                  "type": "string",
                  "description": "The task ID to cancel"
              }
          },
          "required": ["task_id"]
      }
  },

• src/kiro_cli_mcp/server.py:121-122 (registration)

  Registration in the MCP call_tool handler dispatch that matches the tool name and invokes the handler function.

  elif name == "kiro_task_cancel":
      result = await _handle_task_cancel(task_manager, arguments)