Tldraw is a remote MCP server that gives an AI agent 6 tools, among them _exec_callback, _get_canvas_state, exec. It answers at https://tldraw-mcp-app.tldraw.workers.dev/mcp over streamable HTTP, and it listed its tools to an anonymous client, so no sign-in was needed for that. Our probes reached it 6 of 6 times since 11 September 2026; the last check, on 24 October 2026, succeeded.
Draw and visually collaborate with your agents on tldraw's canvas.
Description as the publisher wrote it.
| Tool | What it does |
|---|---|
_exec_callback | App-only: widget calls this to resolve a pending exec request. Inputs: channel, execKey, result, error |
_get_canvas_state | App-only: get the latest checkpoint for a canvas by its canvasId. Inputs: canvasId |
exec | Execute JavaScript code on a tldraw canvas. The code runs in the widget with access to the live `editor` instance, helper functions, and normal js. Use the `search` tool first to discover available Editor methods and shape types. Each canva Inputs: code, canvasId |
read_checkpoint | App-only: read shapes from a checkpoint by ID. Inputs: checkpointId |
save_checkpoint | App-only: save shapes to a checkpoint (from user edits). shapesJson and assetsJson must be JSON array strings. Inputs: checkpointId, shapesJson, assetsJson, bindingsJson, canvasId |
search | Search the tldraw Editor API spec by writing JavaScript that receives a `spec` object and returns a result. The spec contains: spec.members (all Editor methods/properties with name, kind, signature, description, category), spec.categories ( Inputs: code |
Every block below points at https://tldraw-mcp-app.tldraw.workers.dev/mcp. Pick your client and copy.
claude mcp add --transport http tldraw https://tldraw-mcp-app.tldraw.workers.dev/mcp
{
"mcpServers": {
"tldraw": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://tldraw-mcp-app.tldraw.workers.dev/mcp"]
}
}
}{
"mcpServers": {
"tldraw": {
"url": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}{
"servers": {
"tldraw": {
"type": "http",
"url": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}{
"context_servers": {
"tldraw": {
"url": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}{
"mcpServers": {
"tldraw": {
"serverUrl": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}{
"mcpServers": {
"tldraw": {
"url": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}name: Tldraw
version: 0.0.1
schema: v1
mcpServers:
- name: Tldraw
type: streamable-http
url: https://tldraw-mcp-app.tldraw.workers.dev/mcp{
"mcpServers": {
"tldraw": {
"httpUrl": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}{
"mcpServers": {
"tldraw": {
"type": "http",
"url": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}{
"mcpServers": {
"tldraw": {
"url": "https://tldraw-mcp-app.tldraw.workers.dev/mcp"
}
}
}goose://extension?url=https%3A%2F%2Ftldraw-mcp-app.tldraw.workers.dev%2Fmcp&type=streamable_http&id=tldraw&name=Tldraw
It answered our last check on 24 October 2026 in 252 ms. Over 6 checks since 11 September 2026 it answered 6. Status here comes from Neuronto's own probes, not from the publisher.
6, as returned by the server's own tools/list on 22 September 2026: _exec_callback, _get_canvas_state, exec, read_checkpoint, save_checkpoint, search.
It answered our last check on 24 October 2026 in 252 ms. Over 6 checks since 11 September 2026 it answered 6.
Put the server under mcpServers in ~/.cursor/mcp.json with the URL https://tldraw-mcp-app.tldraw.workers.dev/mcp. The exact JSON is above, with the same setup for Claude Code, Claude Desktop, VS Code, Windsurf, Zed, Gemini CLI and the rest.
It listed its tools to an anonymous client, so reading the tool list needs no key. Individual tools can still ask for credentials in their own inputs.
Publisher: io.github.tldraw. Endpoint host: tldraw-mcp-app.tldraw.workers.dev. ARD identifier: urn:air:io.github.tldraw:mcp:tldraw.
First indexed 10 September 2026. The tools are exactly what the server returned; nothing on this page
is inferred from its name. Ranking in Neuronto search is never for sale.