I built a browser agent that plays a game by looking at pixels

A lot of "AI agents" are just an LLM with a REST API stitched to it. I wanted to build something where the agent can't cheat, no DOM shortcuts, no hidden game state. Just a browser, a screen, and a mouse. So I built a small agent that plays Block Champ on CrazyGames. Repo: Here's what was interesting about it, and the techniques all transfer to any canvas-based web app (whiteboards, in-browser IDEs, design tools, games). The problem with "browser agents." Most browser automation tutorials assume the app cooperates: meaningful HTML, readable labels, queryable buttons. Games don't.