// SCAN REPORT

Gcp

17 tools. 4 can modify or destroy data without limits.

4 write tools that can modify data. Rate limits recommended.

Last updated: 29 Jun 2026

4 can modify or destroy data

13 read-only

17 tools total

Community server · catalogue entry verified 29/06/2026

How to control Gcp ↓

TOOL MAP

What Gcp exposes to your agents

Read (13) Write / Execute (4) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous Gcp tools

Execute · High start_instance This tool executes a command that changes infrastructure state by starting a cloud compute resource. Execute · High stop_instance This tool triggers a state-changing operation on cloud infrastructure. Execute · High publish_message Publishing a message to a Pub/Sub topic triggers external operations and downstream consumers. Execute · High query_bigquery Executing arbitrary SQL can read, modify, or destroy data depending on the query (SELECT, INSERT, UPDATE, DELETE, DROP, etc.).

4 of Gcp's 17 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Gcp

PolicyLayer is an MCP gateway — it sits between your AI agents and Gcp, and nothing reaches the server without passing your rules. These are the rules we recommend:

Cap read operations

{
  "get_instance": {
    "limits": [
      {
        "counter": "get_instance_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Gcp — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON GCP →

Instant setup, no code required.

FULL CATALOGUE

All 17 Gcp tools

EXECUTE 4 tools

Execute publish_message Publish a message to a Pub/Sub topic. Execute query_bigquery Execute a BigQuery SQL query. Execute start_instance Start a stopped Compute Engine instance. Execute stop_instance Stop a running Compute Engine instance.

READ 13 tools

Read get_instance Get details for a single Compute Engine instance. Read get_object_metadata Return metadata for a single Cloud Storage object. Read get_project Fetch metadata for a single project. Read list_buckets List Cloud Storage buckets in a project. Read list_datasets List BigQuery datasets in a project. Read list_instances List Compute Engine instances. Read list_objects List objects (blobs) in a Cloud Storage bucket. Read list_projects List GCP projects visible to the active credentials. Read list_service_accounts List IAM service accounts in a project. Read list_subscriptions List Pub/Sub subscriptions in a project. Read list_tables List BigQuery tables in a dataset. Read list_topics List Pub/Sub topics in a project. Read read_logs Read log entries matching an advanced log filter.

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Gcp

Is the Gcp MCP server safe to use without restrictions? +

The Gcp server is primarily read-only with 13 read tools. While it cannot modify data, an agent in a retry loop can make thousands of API calls per minute, exhausting rate limits and running up costs. Rate limiting is still recommended.

How many tools does the Gcp MCP server expose? +

17 tools across 3 categories: Execute, Read, Write. 13 are read-only. 4 can modify, create, or delete data.

How do I enforce a policy on Gcp? +

Register the Gcp MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Gcp tool call.

Deterministic rules across all 17 Gcp tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON GCP →

Instant setup, no code required.

17 Gcp tools catalogued and risk-classified — across an index of 43,000+ MCP servers.