zouhirdx

Because I find it difficult to read the textual output of agents after each modification or planning, and relate everything to the big picture, I built this simple application (with the help of Claude Code).

It is an Electron application that has an integrated terminal linked to the Chalkboard. Any agent you run in this terminal can use it and express its textual outputs in graphics that are easier to review.

Some may find the sounds a little annoying, but it's just an adjustable option, which I prefer.

Using it may cause a bit more token consumption or subscription quota, but in my experience with Claude Code I haven't seen a noticeable difference.

In general, the project is open source for anyone who wants to try it or modify it.

I really wanted to try it, but I couldn't find Ollama Cloud among the options. Most of my subscriptions are official Claude and Codex subscriptions, not API key accounts.

I have great respect for the amount of effort that went into this user experience, even though I haven't seen it from the inside yet.

Thanks! Sorry about the slots, I just raised the limit and idle desktops now free up after 10 minutes, so it should be easier to get in.

It's not a display server. Each app's frontend is a normal web page that runs in your browser, in its own sandboxed frame, so typing, scrolling, menus and dragging are all local, just like any web app. The desktop itself (windows, taskbar, the map) also runs in the browser.

What goes over the network is only what the app decides to send: messages and events between its frontend and its backend on the server, over one WebSocket. So latency shows up where any web app would feel it, when it waits for data from its server, not on every frame or keystroke.

The one exception is something like the Terminal, where every keystroke goes to a shell on the host, so it feels like SSH on a slow link.