Excalidraw MCP Server
Allows AI agents to run a live Excalidraw canvas and draw diagrams programmatically.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Excalidraw MCP ServerDraw a flowchart for the login process"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Excalidraw MCP Server & Agent Skill
Let your AI agent run a live Excalidraw canvas and draw diagrams. This repo provides:
MCP Server: Connect via Model Context Protocol (Claude Desktop, Cursor, Claude Code, etc.)
Agent Skill: Portable skill for OpenClaw agents and other skill-enabled agents
Built on top of mcp_excalidraw by @yctimlin. Added shareable links, image export, agent skill, and a hosted API.
Demo
Related MCP server: Excalidraw MCP App Server
Hosted Instance
A public instance is running at https://excalidraw-mcp.up.railway.app — no setup needed.
{
"mcpServers": {
"excalidraw": {
"url": "https://excalidraw-mcp.up.railway.app/mcp"
}
}
}Self-Hosting
Local
git clone https://github.com/0xArtex/excalidraw-mcp.git
cd excalidraw-mcp
npm install
npm run build
npm run canvasServer starts at http://localhost:3000.
Docker
docker build -f Dockerfile.canvas -t excalidraw-mcp .
docker run -p 3000:3000 -e PUBLIC_URL=https://your-domain.com excalidraw-mcpEnvironment Variables
Variable | Default | Description |
|
| Server port |
| auto-detected | Public URL for edit links |
|
| Internal URL for renderer (auto-resolved) |
|
| Rate limit per IP per minute |
|
| Max elements per render request |
|
| Render timeout (5 min) |
Configure MCP Clients
The MCP server can be connected via Streamable HTTP (hosted) or stdio (local).
Claude Desktop
Hosted:
{
"mcpServers": {
"excalidraw": {
"url": "https://excalidraw-mcp.up.railway.app/mcp"
}
}
}Local (node):
{
"mcpServers": {
"excalidraw": {
"command": "node",
"args": ["/path/to/excalidraw-mcp/dist/index.js"],
"env": {
"EXPRESS_SERVER_URL": "http://localhost:3000"
}
}
}
}Claude Code
claude mcp add excalidraw --scope user \
-e EXPRESS_SERVER_URL=http://localhost:3000 \
-- node /path/to/excalidraw-mcp/dist/index.jsCursor
Config location: .cursor/mcp.json
{
"mcpServers": {
"excalidraw": {
"url": "https://excalidraw-mcp.up.railway.app/mcp"
}
}
}Agent Skill
For OpenClaw and other skill-enabled agents, install the skill to render diagrams with a single API call — no MCP client needed.
The skill is available at skills/excalidraw-canvas/ or can be installed from ClawhHub.
See SKILL.md for usage — it's a single curl call.
MCP Tools
Tool | Description |
| Start a new diagram session, returns live canvas URL |
| Create a shape (rectangle, ellipse, diamond, arrow, text, line) |
| Create multiple elements at once |
| Remove an element |
| Finalize and get shareable link with image |
Development
npm run dev # Watch + hot reload
npm run build # Build everything
npm run type-check # TypeScript checkLicense
MIT
Acknowledgments
@yctimlin - Original mcp_excalidraw project
Excalidraw - The drawing library
Model Context Protocol - The MCP specification
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/0xArtex/excalidraw-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server