MCP (Model Context Protocol) server for managing local tldraw canvas files (.tldr).
Existing tldraw MCP servers let AI draw on an in-memory canvas. This project is different β it reads, writes, and searches local .tldr files on disk, making tldraw a persistent visual scratchpad that AI agents can programmatically update.
- π Read β Load and parse
.tldrfiles - βοΈ Write β Create and update canvas files with validation
- π List β Enumerate all
.tldrfiles with metadata - π Search β Full-text search across all canvases
- π· Shape CRUD β Add, update, delete shapes programmatically
npx @talhaorak/tldraw-mcpgit clone https://github.com/talhaorak/tldraw-mcp.git
cd tldraw-mcp
bun install
bun run startAdd to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"tldraw": {
"command": "npx",
"args": ["@talhaorak/tldraw-mcp"],
"env": {
"TLDRAW_DIR": "/path/to/your/tldraw/files"
}
}
}
}Add to your OpenClaw config:
mcp:
servers:
tldraw:
command: npx
args: ["@talhaorak/tldraw-mcp"]
env:
TLDRAW_DIR: /path/to/your/tldraw/filesOr install as a skill:
curl -fsS https://raw.githubusercontent.com/talhaorak/tldraw-mcp/main/SKILLS.md | bash| Variable | Default | Description |
|---|---|---|
TLDRAW_DIR |
~/.tldraw |
Base directory for .tldr files |
Read a .tldr file and return its parsed content.
tldraw_read({ path: "notes.tldr" })
Write or update a .tldr file (validates format).
tldraw_write({
path: "notes.tldr",
content: { /* tldraw file object */ }
})
Create a new empty tldraw canvas.
tldraw_create({ path: "new-canvas.tldr", name: "My Canvas" })
List all .tldr files with page/shape counts.
tldraw_list({ recursive: true })
Search text content across all canvases.
tldraw_search({ query: "TODO", searchIn: "text" })
Get all shapes from a file, optionally filtered by page.
tldraw_get_shapes({ path: "notes.tldr", pageId: "page:abc123" })
Add a new shape to a canvas.
tldraw_add_shape({
path: "notes.tldr",
shape: {
type: "geo",
x: 100,
y: 100,
props: {
w: 200,
h: 100,
geo: "rectangle",
color: "blue",
fill: "semi"
}
}
})
Update properties of an existing shape.
tldraw_update_shape({
path: "notes.tldr",
shapeId: "shape:abc123",
updates: { x: 200, props: { color: "red" } }
})
Delete a shape from a canvas.
tldraw_delete_shape({
path: "notes.tldr",
shapeId: "shape:abc123"
})
- Visual scratchpad for AI agents β AI updates a canvas you can view in tldraw
- Diagram generation β Create flowcharts, architecture diagrams programmatically
- Note organization β Search and organize visual notes across multiple canvases
- Integration with tldraw desktop/VS Code β Files sync automatically
This server works with tldraw v2 format:
{
"tldrawFileFormatVersion": 1,
"schema": {
"schemaVersion": 2,
"sequences": { ... }
},
"records": [
{ "id": "document:document", "typeName": "document", ... },
{ "id": "page:xxx", "typeName": "page", ... },
{ "id": "shape:xxx", "typeName": "shape", "type": "geo", ... }
]
}# Install dependencies
bun install
# Run in development mode
bun run dev
# Type check
bun run typecheck
# Build for distribution
bun run build
# Run tests
bun test./scripts/publish.sh # Auto-increment patch (0.1.0 β 0.1.1)
./scripts/publish.sh 0.2.0 # Use specific version
./scripts/publish.sh minor # Bump minor (0.1.1 β 0.2.0)
./scripts/publish.sh major # Bump major (0.2.0 β 1.0.0)The script updates package.json, commits, tags, and pushes. GitHub Actions handles npm publish automatically via Trusted Publishers.
- Path traversal prevention β Relative paths can't escape
TLDRAW_DIR - Format validation β All writes are validated against tldraw schema
- No network access β Purely local file operations
MIT Β© Talha Orak