What it does

Queries Pulumi stacks: current resources, outputs, and previews. Infrastructure is the category where an unattended agent is most dangerous, because the blast radius is everything. Use it to answer questions and to generate a preview, then read the preview yourself.

Install

Terminal

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

Use it for

Reading stack outputs and resource state before changing infrastructure

Skip it if

Letting an agent apply changes unattended

When to use it

Reach for this when the task needs infrastructure 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

Pulumi MCP