GCP MCP

Instructions

    Start a Compute Engine instance.
    
    Args:
        project_id: The ID of the GCP project
        zone: The zone where the instance is located (e.g., "us-central1-a")
        instance_name: The name of the instance to start
    
    Returns:
        Status message indicating whether the instance was started successfully
    

Implementation Reference

• src/gcp_mcp/gcp_modules/compute/tools.py:207-242 (handler)

  The core handler implementation for the 'start_instance' tool. It uses the Google Cloud Compute Engine API to start a specified VM instance, polls the operation status until completion, and returns success or error message. The @mcp.tool() decorator registers it with the MCP server using the function name as the tool name and type hints/docstring for schema.

  def start_instance(project_id: str, zone: str, instance_name: str) -> str:
      """
      Start a Compute Engine instance.
      
      Args:
          project_id: The ID of the GCP project
          zone: The zone where the instance is located (e.g., "us-central1-a")
          instance_name: The name of the instance to start
      
      Returns:
          Status message indicating whether the instance was started successfully
      """
      try:
          from google.cloud import compute_v1
          
          # Initialize the Compute Engine client
          client = compute_v1.InstancesClient()
          
          # Start the instance
          operation = client.start(project=project_id, zone=zone, instance=instance_name)
          
          # Wait for the operation to complete
          operation_client = compute_v1.ZoneOperationsClient()
          
          # This is a synchronous call that will wait until the operation is complete
          while operation.status != compute_v1.Operation.Status.DONE:
              operation = operation_client.get(project=project_id, zone=zone, operation=operation.name.split('/')[-1])
              import time
              time.sleep(1)
          
          if operation.error:
              return f"Error starting instance {instance_name}: {operation.error.errors[0].message}"
          
          return f"Instance {instance_name} in zone {zone} started successfully."
      except Exception as e:
          return f"Error starting instance: {str(e)}"

• src/gcp_mcp/server.py:39-39 (registration)

  Top-level registration call in the main MCP server file that invokes the compute module's register_tools function, thereby registering the start_instance tool (among others) with the FastMCP server instance.

  compute_tools.register_tools(mcp)

• src/gcp_mcp/gcp_modules/compute/tools.py:6-6 (registration)

  Module-level registration function that defines and registers all compute tools, including start_instance via nested @mcp.tool() decorators.

  def register_tools(mcp):

• src/gcp_mcp/gcp_modules/compute/tools.py:207-218 (schema)

  Input schema defined by function parameters (project_id: str, zone: str, instance_name: str) and output str, along with detailed docstring describing args and return value, used by MCP for tool schema generation.

  def start_instance(project_id: str, zone: str, instance_name: str) -> str:
      """
      Start a Compute Engine instance.
      
      Args:
          project_id: The ID of the GCP project
          zone: The zone where the instance is located (e.g., "us-central1-a")
          instance_name: The name of the instance to start
      
      Returns:
          Status message indicating whether the instance was started successfully
      """