What it does

Exposes gcloud operations to an agent: listing projects, inspecting resources, and running read commands against Google Cloud APIs. Best used against a sandbox project first, because the same credentials that read can also act. The tool catalog is broad, which means the first call is usually the model guessing at the right subcommand.

Install

Terminal

npx -y @google-cloud/gcloud-mcp

Client configuration

Paste this into your MCP client config (Claude Desktop, Cursor, or any client that reads mcpServers), then restart the client.

mcp.json

{
  "mcpServers": {
    "google-cloud": {
      "command": "npx",
      "args": [
        "-y",
        "@google-cloud/gcloud-mcp"
      ]
    }
  }
}

Use it for

Projects already on GCP where the agent should read real resource metadata

Skip it if

Cross-cloud work, or anyone without a billing-enabled project

When to use it

Reach for this when the task needs gcp inside the agent loop. Skip it if a plain shell command or a one-off script is enough — every server you add costs context and adds a failure mode.

Other TypeScript servers

Filesystem — Gives an agent scoped read, write, and search access to directories you explicitly allow
Memory — A persistent knowledge graph stored in a local JSON file, exposed through entities, relations, and observations
Sequential Thinking — Gives the model a scratchpad for structured, revisable reasoning: numbered thoughts that can branch, revise, and conclude

Google Cloud MCP