Area · 3 projects
Games
Can an agent play, test and build games the way people experience them: on the game's clock, next to human players, and without being told the rules?
Agents can already generate a playable game and keep improving it. The projects here go after what that loop does not measure. As in CUA-SWE, the agent sees only the screen and uses only keyboard and mouse, and the verifier reads the game's own state.
Projects
Each project is a direction with room for many kinds of task. Its page says what it asks and what a verifier there should establish.
Game development loop
Start hereAgents change a game, or the engine under it, and check the result by playing it the way a player would.
Can an agent find out by playing what is wrong with a game, fix it at the right level, and confirm the fix the way a player would?
0 tasksOpen the project →Live play with people
Main betThe agent plays a live game with or for people, a teammate or a whole crowd, seeing only the screen and working out from it what they want.
Can an agent act in a live game while working out what people want, from one teammate's cues to a crowd's chat?
0 tasksOpen the project →Unknown rules
Training environmentNobody tells the agent the rules, of a game level or of an unfamiliar piece of software. It should probe cheaply, rule out what it sees is false, and act once it knows.
Does the agent gather information before it acts, and keep and prune hypotheses about rules nobody told it?
0 tasksOpen the project →Propose a project
Have a question in Games that current benchmarks miss? Propose it as a project: the question, what tasks could look like, and how they could be verified.
opens a GitHub issuePropose →Tasks in this area
Claims, open pull requests and merged tasks across its projects, from the board.
Proposed projects
Open proposals for this area. Comment on the issue to support or shape one.