Enable Monk MCP
Every agent connection starts the same way. Open your project in an IDE where Monk is installed, then run the command palette (Cmd+Shift+P / Ctrl+Shift+P):

Coming soon: Headless mode — no IDE required. Plug Monk directly into Claude Code, Codex, Gemini CLI, or any other CLI agent without running a separate IDE window. Upvote it on our roadmap.
Pick Your Agent
Each agent needs one more step on its side to trust the Monk MCP server. Follow the guide for your agent:VS Code / GitHub Copilot
Enable Monk tools in Copilot chat
Cursor
Approve Monk tool usage in Cursor
Windsurf
Enable MCP tools in Cascade
Claude Code
Project-local MCP config for Claude Code
Codex
Codex CLI or the Codex IDE extension
Gemini CLI
Project-scoped settings for Gemini
Antigravity
MCP management UI in Antigravity
Two Patterns
IDE-embedded agents like Copilot, Cursor, Windsurf, and Antigravity run inside the same IDE as Monk. You install Monk, enable MCP, and allow the agent to use Monk tools — all in one place. Standalone agents like Claude Code, Gemini CLI, and Codex run in a separate terminal. Keep Monk open in your IDE with the project loaded, then launch the standalone agent in the same directory. Monk writes the config file the agent needs automatically.What Monk Configures
When you enable MCP, Monk writes client-native config for the current project:| Agent | Config location |
|---|---|
| VS Code / Copilot | .vscode/mcp.json |
| Cursor / Windsurf | .cursor/mcp.json |
| Claude Code | ~/.claude.json (project-local entry) |
| Gemini / Antigravity | .gemini/settings.json |
| Codex | .codex/config.toml |
Next Steps
First deployment
Deploy your first application through your agent
MCP support details
How Monk MCP works under the hood

