Let an AI play-test your game through a text harness
A developer wrapped a browser game so an AI could run playtests, reproduce bugs, and flag problems before each milestone.
- What they did
- The developer built a Node.js layer around a browser-based crossword dungeon without changing the game files. That wrapper added a text view, synthetic key presses, and an HTTP interface, so the agent could interact with the game through curl requests. They also had it create fixtures for bugs and told it to test each milestone feature before advancing.
- How it went
- The agent got through milestone 13, which added five special encounters, in 12 minutes and used about 120k tokens. It found a cage bug and fixed it by itself. Manual playtesting was still useful and turned up more issues, but it took less than 30 minutes.
- Tips
- This works better for games driven by text and turns than for browser games that depend on screenshots. Keep a human pass in the loop, since the agent can miss some problems even when it does well.
Try it yourself with Claude Code
I have a game at [path]. Build a small test harness around it without changing the game files if possible. Expose the current game state as text or JSON, accept actions such as [moves/keys/commands], and let an AI agent play through [number] sessions. After each milestone or feature set, stop and wait for my approval before continuing to the next stage. For every issue you find, give a clear reproduction path, note any balance or usability problems, and separate anything you fixed from anything that still needs manual review. Finish with a plain play-test report.
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
Let a coding agent enter your game world
A developer connected their coding agent to Overlands through a separate client so it could appear inside the game as a player.
Run a Micropolis city with an agent
Hook an agent to a headless Micropolis simulator and let it manage a city from a simple goal sheet.
Build a bot for a real-time strategy game
They used LLM-written code to control bots in strategy-game benchmarks and had them compete in tournaments and ladders.
Use an AI to referee a D&D 5e dungeon
They ran a browser D&D crawl with three characters while the agent checked rules text and monster stats from separate sources.
Let an agent pick and play a simple game
AI Village let agents choose their own games, and one agent spent the week on Minesweeper while also keeping a shared score sheet.
Use Codex to shape game architecture and animation blending
A game builder had Codex handle core architecture and improve animation transitions by reading engine docs and wiring in merging logic.