翻訳待ち：Show HN: AI Game Playtester

AI サービスが一時的に利用できないため、復旧後に翻訳を補完します。

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. Download Ziva freeSee Pro plans Free to install · Playtest Agent needs Pro ($50/mo) · Godot 4.2+ today · Other engines via Enterprise 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. 01 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. 02 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. 03 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. 04 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. A real run, unedited: the game only moves while an action runs, so the agent has all the time it needs to think between moves. 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. CapabilityCursor · Claude Code · CopilotZiva 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 AI game testing, answered What is AI game testing? AI game testing is QA done by an agent that plays your game instead of a person. It sends real keyboard, mouse and gesture input, watches the screen and the running scene, and reports whether a feature works. Ziva's Playtest Agent does this for Godot games: give it a goal, and it plays until it can return a pass or fail verdict backed by what it measured. Can AI playtest a game automatically? Yes. You give Ziva a suite of tests written in plain English, like 'the score goes up when the player clears a pipe'. The agent reads your code to learn how the game takes input, launches it in a sandbox, plays until each test is settled, and records a video of the whole run. Dozens of instances can run in parallel, so a full test suite finishes in minutes. How do you playtest a game? Traditionally you recruit players, watch them play, take notes, and repeat every build — studios collectively spend around $1.7B a year doing this. With AI playtesting you describe what should happen, and an agent plays the game start to finish on every build, so human playtests can focus on feel and fun instead of catching broken buttons. What tools can automate video game testing? Generic UI-automation tools break on games because games render one canvas with no buttons to select. Game-aware options are engine test frameworks for unit tests, record-and-replay input scripts, and AI agents that actually play. Ziva's Playtest Agent is the AI option for Godot: no test scripts to write or maintain, just goals in plain English. How is AI playtesting different from PlaytestCloud or human playtesters? Services like PlaytestCloud recruit real people, which takes days and costs per tester, and is the right tool for judging fun and first impressions. AI playtesting answers a different question — does the game actually work — in minutes, on every build, without recruiting anyone. Most studios need far fewer human sessions once an agent catches the functional breaks first. Does AI playtesting send my game to the cloud? Ziva's playtests run fully locally: your game launches and is driven on your own machine, and your project files never leave it. The only remote calls are to zero-data-retention LLMs for thinking and image analysis, so nothing about your game is stored or trained on. How much does AI game testing cost? Ziva is free to download and the Hobby tier costs nothing, but the Playtest Agent needs Pro at $50/month. A full playtest runs an agent for the length of a game session, so it uses far more tokens than a normal chat turn. Which engines and platforms does the Playtest Agent support? Any engine. Godot 4.2 and later is self-serve today, Unity runs through Ziva's Unity editor bridge, and studios on Unreal or a custom engine get set up through an Enterprise inquiry. Playtests cover games built for PC, mobile, and console. Ready to accelerate your game development? Start for free. Join hundreds of game developers who are building faster with AI-powered assistance. Building for a studio? Contact us for Enterprise options