Full game tester
Game Studios spent $1.7B a year on playtesting. Don't be one of them.
Fully playtest your game with AI in minutes: Ziva's playtest agent is able to fully complete games, can run dozens of instances in parallel, and works for PC, mobile, and console.
One real run on Godotcraft, start to verdict:
- 1The playtest QA agent is told to chop down a tree and make a crafting bench.
- 2It analyzes the game, identifying movement, pathfinding, tree blocks, and crafting.
- 3It walks towards the tree, chops a log down, picks it up, and creates the Crafting Bench.
- 4A video showcasing the playtest start to finish is analyzed by AI for pass/fail, and saved for viewing.
How it works
AI driven playtesting
You give it a suite of tests to run, like “the score goes up when the player clears a pipe”, and the AI plays until it sees that happen.
Reads your code first
Analyzes the game code to figure out how to accomplish the goal, and where to add temporary hooks + logging to execute.
Launches it, then freezes it
LLMs don't work in realtime; we launch your game in a sandbox and pause it, giving time for the LLM to think.
Plays, and measures every move
Sends real inputs to the game, and has output measured by screenshots, custom hooks that receive queries from the LLM, and added logging.
Ends with a verdict
The AI looks at everything completed as a whole to determine if the tasks were completed successfully, and saves a video for you to review as well.
Playtests run fully locally on your machine. The only remote calls are to zero-data-retention LLMs for thinking and image analysis - your project never leaves your computer.
Versus a general AI agent
Most AI agents can't play your game
Cursor, Claude Code and Copilot are great at writing the code. None of them can pick up the controller. When one of them says “I fixed it”, it is reading the source and guessing, and you find out otherwise when you play the game yourself.
| Capability | Cursor · Claude Code · Copilot | Ziva Playtest Agent |
|---|---|---|
| Writes and runs your unit tests | via the terminal | run_tests, headless |
| Launches and plays the actual game | no hands on the game | drives it turn by turn |
| Sends real player input | can't reach the window | keys, clicks, drags, aim |
| Measures the live scene while playing | only reads the code | probes + your own signals |
| Cannot edit what it is judging | edits it, then declares success | ships with no write tools |