Coding agents
GitHub Copilot
Add the Stableship MCP server to GitHub Copilot, in Copilot CLI and VS Code, with a token.
GitHub Copilot connects to the Stableship MCP server with a token, in Copilot CLI and in VS Code, and reads .github/copilot-instructions.md and AGENTS.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.
Note
We have not yet tested GitHub Copilot with Stableship end to end. It connects over MCP with the steps below, taken from GitHub Copilot's own docs. If a step does not match what you see, tell us at support@stableship.ai.
Connect
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 this to ~/.copilot/mcp-config.json, which Copilot CLI and VS Code both read:
{ "mcpServers": { "stableship": { "type": "http", "url": "https://api.stableship.ai/mcp", "headers": { "Authorization": "Bearer sst_your_token" }, "tools": [ "*" ] } }}Keep the token out of your repository.
Check it works
Start a new Copilot chat or session. Then ask GitHub Copilot to list your Stableship projects: it calls list_projects and names your app.
If it does not connect:
- Check the token was pasted whole.
- Run
/mcpin Copilot CLI: Stableship should be listed.
Keep the rules in your repository
Ask Copilot 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 .github/copilot-instructions.md. Stableship never writes to your repository, so your agent commits these.
For your coding agent: this page as Markdown