Provides tools for managing Google Cloud Batch jobs and Compute Engine VM instances, including creating, listing, inspecting, starting, stopping, resetting, and deleting VMs, as well as submitting and managing batch jobs.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Google Cloud Batch (HPC) & Compute Engine VM MCP Server Submit a batch job to run my Python script across 10 parallel tasks."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server that empowers AI assistants (Antigravity CLI, IDEs, Claude, etc.) to manage Google Cloud Batch (HPC / Batch computing workloads) and Google Compute Engine Virtual Machines (VMs) directly through conversational prompts.
submit_batch_script_job: Submit bash/shell script batch computing jobs (configurable CPU millicores, RAM MiB, task counts, parallelism, and Spot/Preemptible VMs).
submit_batch_container_job: Submit containerized batch workloads with Docker/OCI images.
list_batch_jobs: List batch jobs in a project/region with current states (QUEUED, RUNNING, SUCCEEDED, FAILED).
get_batch_job: Inspect job details, task statuses, failure reasons, and direct Google Cloud Console links.
cancel_batch_job: Cancel a running or scheduled batch job.
delete_batch_job: Permanently remove a batch job definition.
list_batch_tasks: List individual tasks and exit codes inside a job group.
create_gcp_vm: Provision new VM instances in Google Cloud Console with customized machine types (e.g. e2-medium, c2-standard-8, a2-highgpu-1g), OS families (Ubuntu, Debian, COS, CentOS), boot disk size/type, startup scripts, network tags, and Spot pricing.
list_gcp_vms: List instances across one or all zones with runtime status, machine type, internal IP, and external IP.
get_gcp_vm: Retrieve detailed hardware, networking, and disk specifications for an instance.
start_gcp_vm: Start a stopped VM instance.
stop_gcp_vm: Gracefully stop a running VM instance.
reset_gcp_vm: Hard reset a VM.
delete_gcp_vm: Terminate and delete a VM instance.
get_vm_serial_port_output: Inspect serial console boot logs (crucial for debugging startup scripts and HPC initialization).
check_gcp_connection: Verify active GCP credentials, service account email, and default project/region/zone.
Related MCP server: GCP MCP
The server automatically detects credentials in the following order:
Environment Variable: GOOGLE_APPLICATION_CREDENTIALS pointing to a Service Account key JSON file.
Local Key File: Place your Google Cloud Service Account JSON file as gcp_credentials.json directly in this folder (C:\Users\banda\gcp-batch-mcp-server\gcp_credentials.json).
Application Default Credentials (ADC): Run gcloud auth application-default login in your terminal.
Ensure your Service Account or user identity has:
Batch Job Administrator (roles/batch.jobsAdmin)
Compute Instance Admin (v1) (roles/compute.instanceAdmin.v1)
Service Account User (roles/iam.serviceAccountUser)
.env)Copy .env.example to .env and fill in your values:
GCP_PROJECT_ID=your-project-id
GCP_REGION=us-central1
GCP_ZONE=us-central1-a
GOOGLE_APPLICATION_CREDENTIALS=C:\Users\banda\gcp-batch-mcp-server\gcp_credentials.jsoncd C:\Users\banda\gcp-batch-mcp-server
.venv\Scripts\python.exe src\server.pycd C:\Users\banda\gcp-batch-mcp-server
.venv\Scripts\python.exe -m unittest discover -s tests -p "test_*.py"This server cannot be deployed
Provides capabilities that let LLM agents perform a range of infrastructure management tasks.
The Google Compute Engine MCP server is a fully-managed Model Context Protocol server that provides tools to manage Google Compute Engine resources through AI agents. It enables capabilities including instance management (creating, starting, stopping, resetting, listing), disk management, handling instance templates and group managers, viewing machine and accelerator types, managing images, and accessing reservation and commitment information. The server operates as a zero-deployment, enterprise-grade endpoint at https://compute.googleapis.com/mcp with built-in IAM-based security.
Interact with your Google Cloud Composer resources using natural language commands.
Interact with your Google Bigtable resources using natural language commands.