Skip to content

Integration · MCP

Connect Gemini CLI to Archboard

With the Archboard MCP server, Gemini CLI can list, read, create and edit your diagrams. Its changes appear live on the canvas, next to its own cursor labeled with your name.

MCP server URLhttps://getarchboard.app/mcp

About Gemini CLI

Gemini CLI is Google’s open-source AI agent for the terminal.

Set up Gemini CLI

  1. Add Archboard to Gemini CLI

    Run this in your terminal:

    gemini mcp add --scope user --transport http archboard https://getarchboard.app/mcp

    This app needs a business subscription or a paid API key from its maker.

    When your browser opens, sign in to Archboard and approve the app.

    If the browser doesn’t open, type this in the app:

    /mcp auth archboard

    The command adds Archboard for your user, so it is available in every folder.

  2. Sign in and choose access

    Your app opens Archboard. Sign in with Google or GitHub, pick the organizations and workspaces it can use, choose read, edit, or delete for each, and give it a name.

  3. Ask for a diagram

    Ask your agent to draw or update a diagram, then open it in Archboard to watch the changes arrive live.

Your AI app needs to support remote MCP servers with OAuth sign-in.

Try these prompts

Once connected, ask Gemini CLI in plain words, for example:

  • “Draw our checkout service on Kubernetes with Postgres and Redis.”
  • “Add a cache next to checkout-api in my Payments diagram.”
  • “Undo your last change.”
Example prompts and the diagrams they draw

Plan limits

MCP works on every plan, including Free, with the same plan limits as the editor. Agent tool calls also have a per-minute rate limit. On Free you edit live with your AI agents; editing live with teammates needs Pro or Enterprise seats.

  • Free: 3 diagrams and 15 diagram nodes + 150 notes, text, drawings, stickers & mind map topics per diagram.
  • Pro and Enterprise: unlimited diagrams and nodes, plus seats for teammates to edit live.
View pricing

Troubleshooting

The Archboard sign-in page doesn’t open

Use the sign-in command shown in the setup above, if there is one. Otherwise remove Archboard from the app and add it again: the app opens the sign-in page when it connects.

The agent can read diagrams but can’t change them

If you approved read-only access when connecting, connect the app again and approve edit access. If the app’s permission in that workspace is Read, change it to Edit under Settings → Connected apps; the change applies to its next call.

The agent can’t find a workspace or diagram

An app can use the organizations and workspaces you allowed when you connected it. Change them under Settings → Connected apps, with no reconnect needed.

The agent says a plan limit is reached

MCP has the same plan limits as the editor. When a limit is reached, the agent shows Archboard’s message with an upgrade link. Remove diagrams or nodes you no longer need, or upgrade to Pro.

Calls fail with a rate limit

Agent tool calls have a per-minute rate limit. Wait a moment and try again; asking for fewer, larger changes helps.

The app has no option for custom MCP servers

Some apps allow custom MCP servers on certain plans, or after an admin turns them on.

Open a canvas in a minute

Sign in with Google or GitHub. Free for a single seat. Upgrade when the team joins.