{"data":{"slug":"desktop-commander-mcp","name":"Desktop Commander MCP","tagline":"MIT MCP for terminal, files, and processes on your desktop host","category":"agents","categoryName":"AI Agents","url":"https://github.com/wonderwhy-er/DesktopCommanderMCP","github":"https://github.com/wonderwhy-er/DesktopCommanderMCP","pricing":"open-source","hook":"Desktop Commander MCP gives Claude (and other MCP clients) terminal control, filesystem search/edit, and process management on the host — beyond plain filesystem MCP.","whatsGood":"MIT (wonderwhy-er/DesktopCommanderMCP). Free via npm `@wonderwhy-er/desktop-commander`, npx, Smithery, or Docker isolation path. Extends MCP filesystem server with search/replace editing and long-running terminal commands. ★9753 on GitHub.","theCatch":"local blast radius: shell + files + processes on your machine. Remote MCP path (mcp.desktopcommander.app) expands the trust boundary further. Vs playwright-mcp / BrowserSkill: those are browser control; this is host OS control. Guardrails exist; it is not a sandbox unless you choose Docker.","verdict":"MIT host-OS MCP when agents need real terminal/file control — only with explicit trust boundaries; prefer Docker isolation or skip if browser MCP is enough.","addedAt":"2026-09-26","featured":false,"domain":"AI Agents","subSpecialty":"Agents","subcategory":"MCP","tags":["mcp","desktop-app","self-host","permissive","active"],"capabilities":["mcp"],"surfaces":["MCP","Self-host"],"ecosystem":["Multi-platform"],"licenseModel":"OSS/MIT","githubStars":10097,"starsPerWeek":103,"access":"free","archived":false,"repo":{"createdAt":"2024-12-04","pushedAt":"2026-10-18","license":"MIT","archived":false,"checkedAt":"2026-10-19"}},"related":[{"slug":"claude-code","name":"Claude Code","tagline":"Terminal-first coding agent with deep repository context"},{"slug":"cursor","name":"Cursor","tagline":"The AI-first code editor that understands your codebase"},{"slug":"devin","name":"Devin","tagline":"Autonomous AI engineer for defined engineering tasks"},{"slug":"stagehand","name":"Stagehand","tagline":"AI-native browser automation that mixes code and natural language"},{"slug":"browser-use","name":"Browser Use","tagline":"Open-source framework for LLM-driven browser agents"}],"links":{"self":"/api/gems/desktop-commander-mcp","html":"/gems/desktop-commander-mcp","markdown":"/gems/desktop-commander-mcp.md","category":"/api/gems?category=agents","categoryPage":"/stacks/agents"}}