What it does

Manages Runpod GPU pods: create, inspect, and stop. This is infrastructure with a meter running, which makes it the clearest example of a server that should never have unattended write access. Use it to check status and to start work you are watching.

Install

Terminal

npx -y @runpod/mcp-server

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": {
    "runpod": {
      "command": "npx",
      "args": [
        "-y",
        "@runpod/mcp-server"
      ]
    }
  }
}

Use it for

Spinning up a GPU pod and checking its status from an agent

Skip it if

Anything cost-sensitive; pods bill by the second

When to use it

Reach for this when the task needs gpu 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

Runpod MCP