Gemini CLI

Gemini CLI connects to the Stableship MCP server and reads GEMINI.md from your repository. Once it is connected, ask it to fix what Stableship found; it reads each fix brief, fixes the code and asks Stableship to verify the fix.

Connect

1. Add Stableship in your terminal:

   gemini mcp add --scope user --transport http stableship https://api.stableship.ai/mcp

2. Start Gemini CLI and run /mcp auth stableship. Your browser opens. Approve there.

Approving opens app.stableship.ai: sign in, pick the workspace and approve. No token to copy. You can disconnect it any time in Stableship: Settings, then Connections, then Connected apps.

Use a token instead

First make a token: in Stableship open Settings, then Connections, open API tokens and connected apps, and choose Create token with Read findings and rules and Start checks and verify fixes. Copy it: it starts with sst_ and is shown once. Use it where these steps say sst_your_token.

Add it to ~/.gemini/settings.json with a token:

{  "mcpServers": {    "stableship": {      "httpUrl": "https://api.stableship.ai/mcp",      "headers": {        "Authorization": "Bearer sst_your_token"      }    }  }}

Keep the token out of your repository.

Check it works

Start Gemini CLI again: /mcp lists Stableship. Then ask Gemini CLI to list your Stableship projects: it calls list_projects and names your app.

If it does not connect:

• In Gemini CLI, run /mcp auth stableship, then approve.
• Or use a token instead.

Keep the rules in your repository

Ask Gemini CLI to call get_repo_files and commit what it returns: .gate/rules.md, .gate/context.md, a Stableship section in AGENTS.md, and the same section in GEMINI.md, the file Gemini CLI reads. Stableship never writes to your repository, so your agent commits these.