Connect GitHub Copilot to Rafter’s MCP server
Copilot's agent mode in VS Code reads MCP servers from mcp.json. Add Rafter once and Copilot can search your team's shared memory, follow its skills, and write decisions back.
Open your MCP configuration
In VS Code, run MCP: Open User Configuration from the Command Palette. That opens the
mcp.jsonin your user profile, which applies to every workspace.Add Rafter as an http server
Add Rafter under
servers, then save.{ "servers": { "rafter": { "type": "http", "url": "https://app.heyrafter.xyz/api/mcp" } } }The top-level key is "servers" here, not "mcpServers" — VS Code differs from most other clients on this.
Sign in and pick a workspace
VS Code prompts you to authenticate on the first Rafter call and opens the browser. Pick the workspace to connect. The server then shows under MCP Servers — Installed in the Extensions view.
Use it in agent mode
Switch Copilot Chat to Agent mode and try “Check Rafter first: how do we handle migrations?”
MCP tools are available in agent mode; ask mode does not call them.
Your first five minutes
The connection is the easy part. The habit is what pays: make GitHub Copilot check the shared brain before it improvises. Three things to try in Copilot:
- “What do we know about [a client or project]?”Rafter’s search answers from your team’s actual notes — with what they cite.
- “How do we [something a teammate has done before]? Follow the skill if we have one.”Skills are the team’s way of doing it, taught once and reused.
- “Save this as a memory: [a decision you just made].”Written back with add_memory — versioned, searchable, there for the next person.
To make it automatic, paste this line into GitHub Copilot’s project or system instructions:
Check Rafter first: search our skills and memories before you answer, and save anything durable back with add_memory.
A longer startup prompt and a ready-made Claude Code skill live in the rafter-mcp repo.
Troubleshooting
Copilot isn't calling Rafter's tools
Make sure Copilot Chat is in Agent mode — MCP tools are not available in ask mode. The Extensions view, under MCP Servers — Installed, shows whether Rafter is connected and which tools it exposes.
Which workspace should I pick when signing in?
The one whose skills and memories you want this tool to see. A connection sees exactly what you can see — your teams and your role — never more. You can connect again later and grant a different workspace.
Is my team's content used to train AI models?
No. Rafter does not use your content to train generalized or foundation models — see the privacy policy.
Give your whole team the same brain.From $29 per user — see pricing.
Create your account