Cursor knows your repository. It does not know the architecture decision record in a different repository, the vendor PDF, or the meeting where you rejected the approach it is about to suggest. gno mcp install --target cursor wires all of that in as retrieval tools, pinned to an exact runtime and index so the editor opens the same workspace regardless of how it was launched.

Connects via: MCP server
Requires: Bun 1.3+ and one indexed folder
Data boundary: Corpus and index stay on your machine

Install

Connect Cursor in one command.

gno mcp install --target cursor

gno mcp status

gno setup . --apply-profile

What it unlocks

Retrieval across repos, docs, and reference material from inside the editor
Per-collection embedding models, so a code collection can diverge from prose
Structural code chunking for TypeScript, JavaScript, Python, Go, and Rust
A checked-in .gno/index.yml profile so the whole team gets the same retrieval setup

Prove it actually works

A written config file is not a working integration. GNO distinguishes the two, and so should you.

Run gno mcp status to confirm GNO is registered for the target.
Open the Connectors page in gno serve and run the explicit read-only retrieval check. It proves the configured client can actually retrieve, not just that a config file exists.
Ask the client something only your corpus knows. If it answers with a citation into your files, the wiring is real.

Keep reading

Related guides and features.

Cursor setup guide

/docs/cursor

Open

Project profiles

/features/project-setup-profiles

Open

MCP reference

/docs/mcp

Open

All integrations

Every supported client

Browse all