Connect a coding agent to an Excalidraw board
They hooked an installed coding agent to a live Excalidraw canvas so it could read the scene and edit the drawing directly.
- What they did
- They used Drawgent, a single binary that links an existing Claude Code, Codex, or opencode setup to an Excalidraw board. The tool adds canvas controls through MCP, and the person can steer it from a chat panel, by writing AGENT: notes on the board, or by pointing with the laser. The agent checks the current view and a screenshot, then makes live edits and ties each box back to code.
- How it went
- The writeup does not give performance numbers. It does note several limits: macOS builds have not been tested on a real Mac, live Codex attach has not been tested, rooms only mirror one board, and images do not sync.
- Tips
- Make sure you already have an agent installed and signed in before trying this. Rendering depends on Chrome or Chromium, and setup will look for one or install it if needed.
Try it yourself with Cursor
Help me connect [agent name] to [Excalidraw board path or URL] so it can inspect and update the canvas while I work on [project]. First, create or revise an architecture diagram from the code in this repo. Then, when I describe a change like [example change], update the diagram to match. Before you overwrite, delete, send, or submit anything, stop and ask for my approval. When you finish, save the updated diagram back into the repo and explain what changed.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Have coding agents test your onboarding flow
Run coding agents through your product’s onboarding and get a scorecard, fix list, and session traces.
Translate a runtime to Rust module by module
Microsoft used agents to replace the Copilot runtime’s TypeScript with Rust while keeping the implementation’s overall design the same.
Review a feature and pass the findings to an engineer
One developer had an agent inspect a feature for problems, then handed the issue list to an engineer whose agent fixed most of them.
Create a CLI coding agent from a Python LLM base
Claude Code turned Simon Willison’s Python LLM template into a working CLI coding agent from two prompts.
Add simple forgetting to agent memory
A developer added a basic forgetting layer to an agent’s memory and said it worked extremely well right away.
Break one goal into a sequenced coding plan
Ordewell turns a single objective into a step-by-step task list for coding agents, with dependencies, review, and parallel runs.