MCP server · Design tools
CanvasKeep MCP server
by CanvasKeep
Let your AI draw Excalidraw diagrams and save them straight into your CanvasKeep workspace.
This is a small helper that connects your AI assistant to CanvasKeep, a place where you keep Excalidraw diagrams. Once it is set up, you can just describe a diagram in plain words and your AI will draw it and save it for you. It is handy if you sketch systems, flows, or ideas and would rather talk than drag boxes around.
What is an MCP server? The 30-second version
On its own, your AI can only chat. It cannot open your CanvasKeep account or draw anything there. An MCP server is a little helper program that gives your AI one new skill, in this case a connection to CanvasKeep. With this helper running, your AI can look at your workspaces, folders, and diagrams, and create or change them when you ask.
What this MCP server does
You tell your AI what you want, for example a diagram of a web app. Your AI uses this helper to talk to CanvasKeep over the internet. The helper can list your workspaces and folders, create new ones, create a diagram, and fill it with boxes and arrows. If you ask it to, it can also read a diagram you already have or save a new version back. You end up with a real Excalidraw diagram sitting in your CanvasKeep workspace.
What you can do with it
- Draw a new diagram from a plain description, with boxes and labeled arrows
- List your workspaces, folders, and diagrams
- Create a new workspace or folder, including folders inside folders
- Rename or delete a diagram
- Read the full contents of an existing diagram
- Save a changed version of a diagram back to CanvasKeep
- Check a diagram's details and how many elements it has
Try asking your AI
- “Draw an architecture diagram of a web app with a React frontend, a Node API, a PostgreSQL database, and a Redis cache.”
- “Make a new folder called Q3 planning in my main workspace.”
- “List the diagrams in my CanvasKeep workspace.”
- “Read my onboarding flow diagram and tell me how many elements it has.”
What it gives back to you
You get answers in the chat, like a list of your workspaces or folders, or the details of a diagram. When it creates or changes something, it tells you what it did, for example the name of the new diagram. If you ask it to read a diagram, it can describe what is in there. The actual drawing lives in CanvasKeep, ready to open in Excalidraw.
Before you start
What you need
- A CanvasKeep account
- An API key from your CanvasKeep profile menu (a kind of password for apps; CanvasKeep gives you one for free)
- The Claude Code app or another tool that supports MCP servers
- Node.js installed on your computer (the helper runs with a tool called npx)
Good to know
It can delete diagrams for good, so check before you say yes, and keep your API key and master key private.
Install it with your AI
Add CanvasKeep MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check CanvasKeep MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,609 MCP servers here for you.
Who it's for
People who sketch systems, flows, or ideas and would rather describe them in words than draw them by hand.
