Which integrations are available for this server?

Provides tools for interacting with a shared tldraw topology canvas, enabling agents to retrieve topology state and add, update, or delete nodes and edges, as well as clear the topology.

How do I use tldraw Topology MCP?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@tldraw Topology MCP add a node called web-01 and connect it to the load balancer" 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.

tldraw Topology MCP

A shared local tldraw topology service with a persistent HTTP daemon and a thin stdio MCP adapter.

Architecture

daemon.mjs owns the tldraw store, persistence, browser/static HTTP service, sync API, health endpoint, and topology mutations.
server.mjs is a thin stdio MCP adapter. It exposes eight topology tools and forwards calls to the daemon over HTTP.
The MCP adapter does not own the daemon lifecycle. MCP clients can connect and disconnect without destroying the shared canvas state.

Related MCP server: Canvas AIGC

Install and build

npm install
npm run build

Run the long-lived daemon:

npm start

Run the stdio MCP adapter in an MCP host:

npm run mcp

MCP tools

topology_get, node_add, node_update, node_delete, edge_add, edge_update, edge_delete, topology_clear.

Daemon configuration

Variable	Default	Purpose
`TLDRAW_HOST`	`127.0.0.1`	HTTP bind address
`TLDRAW_PORT`	`5173`	HTTP port
`TLDRAW_DATA_DIR`	`<project>/data`	Directory containing `store.json`
`TLDRAW_STORE_FILE`	`<data dir>/store.json`	Exact store path; overrides `TLDRAW_DATA_DIR`
`TLDRAW_DIST_DIR`	`<project>/dist`	Built browser assets
`TLDRAW_CANVAS_URL`	derived local URL	URL reported to MCP clients

The default bind is loopback. External interface exposure belongs to deployment configuration and is not hard-coded in the application.

MCP adapter configuration

Variable	Default	Purpose
`TLDRAW_TOPOLOGY_DAEMON_URL`	`http://127.0.0.1:5173`	Daemon base URL
`TLDRAW_TOPOLOGY_TIMEOUT_MS`	`10000`	Per-call HTTP timeout

HTTP endpoints

GET /healthz — daemon health and current revision.
GET /api/state — complete tldraw document snapshot.
GET /api/changes?since=<revision> — incremental browser synchronization.
POST /api/changes — browser document changes.
POST /api/topology/command — semantic topology command endpoint used by the MCP adapter.

Persistence and concurrency

The daemon is the single writer for the authoritative local document. Runtime state is stored in data/store.json by default and is intentionally excluded from source control. Writes use a temporary file followed by rename. Semantic mutations are serialized so multiple MCP adapters can safely share one daemon.

Service supervision

Operating-system service registration, startup policy, and external supervision are intentionally outside this repository. The daemon is designed to be hosted by a generic supervisor without coupling that supervisor to the MCP adapter.

Version

Current release: v0.1.0.

Upstream

The browser UI is built with the tldraw SDK. See the upstream tldraw project for its licensing and trademark terms.

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityMaintained

ResponsivenessNo issues

Resources

GitHub Repository

Need Help?

Related Servers

Related MCP Connectors

VyrlOAuth
io.github.solarx56
Inspect and edit media canvases, run existing Flows, and retrieve results. Vyrl MCP token required.
Image & Video Processing App Automation
1
DAG Studio WebMCP
io.github.Black-Swan-Causal-Labs
Read and edit a live DAG Studio canvas through its predefined WebMCP tools.
Data Platforms Developer Tools
1
irenictableOAuth
org.benevolabs
Collaborative whiteboard MCP server — create objects, connectors, C4 diagrams, and manage boards
Design Tools Workplace & Productivity Software Architecture
2
Algenta MCP Server A
io.github.thyn-ai
Governed data discovery, exact queries, decisions, simulations, and runtime utilities over MCP.
Data Platforms Search RAG Systems
1402

Related MCP Servers

Timeverse InfCanvas MCP Server
File Systems Image & Video Processing AI & Machine Learning
elimyliu
A
license
Not graded
quality
D
maintenance
Enables AI clients to read and write local infinite canvas data via MCP protocol, with support for image generation and web visualization.
Updated 3 months ago (2026-07-02 07:39 UTC)
MIT
Canvas AIGC
Design Tools AI & Machine Learning Multimedia Processing
3108482409song-commits
A
license
Not graded
quality
B
maintenance
A MCP server that bridges web canvas with local Codex/Claude, exposing 23 tools for reading, creating, modifying, connecting, and generating canvas content via local HTTP bridge and stdio MCP.
Updated 3 months ago (2026-07-17 04:40 UTC)
AGPL 3.0
Nerf (nib) MCP Server
Software Architecture Design Tools
nerfaslol
A
license
Not graded
quality
C
maintenance
Enables AI agents to create and edit diagrams on a live canvas via MCP tools, including adding nodes, edges, and batching operations, with conflict detection via versioning.
Updated a month ago (2026-08-22 20:30 UTC)
70,837 npm
MIT
OpenBoard
Software Architecture Developer Tools
atpaawej
A
license
Not graded
quality
A
maintenance
A local-first whiteboard MCP server that enables AI agents to create, inspect, and update canvas diagrams and shapes collaboratively via 13 semantic tools.
Updated a month ago (2026-08-17 16:58 UTC)
4
MIT