What it does

Controls an Electron application: clicks, keystrokes, and inspection of the running app. It fills a real gap, because desktop apps are invisible to browser automation. Treat it as a QA instrument — it operates whatever window it is attached to, including ones with real data in them.

Install

Terminal

npx -y electron-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": {
    "electron": {
      "command": "npx",
      "args": [
        "-y",
        "electron-mcp-server"
      ]
    }
  }
}

Use it for

Driving a desktop app for QA or a reproducible bug report

Skip it if

Production user machines at scale; this is a test-harness tool

When to use it

Reach for this when the task needs desktop 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

Electron MCP