What problem does it solve? Working with open tldraw Desktop canvases programmatically is impossible without knowing the app's local HTTP server, its per-launch auth token, and its document model. This Skill teaches an agent how to discover the server, authenticate, read and mutate canvas state, and add durable scripted behavior to .tldraw documents. ## Core Features & Use Cases - Canvas inspection and editing: List open documents, read shapes and bindings, and run JavaScript against a live editor through the /api/search and /api/doc/:id/exec endpoints. - Durable document scripts: Create persistent behaviors (clickable UI, animations, custom shapes via config.js) through the /script-workspace endpoint with watcher status verification. - Screenshots and linting: Capture canvas or window screenshots and run helpers.getLints() to validate diagram connections before reporting completion. - Use Case: Ask the agent to arrange and connect boxes on your open project-plan board; it discovers the doc, creates bound arrows between shapes, lints the result, and saves the local document. ## Quick Start Ask the agent to list my open tldraw canvases and add a labeled rectangle to the focused document.