The Dumbest-Looking AI Prompt Just Beat Months of Careful Game-Design Prompt Engineering

Just two days after Claude Opus 5 shipped, AI investor and former HyperWrite CEO Matt Shumer posted a video of a fully playable first-person shooter the model had built entirely on its own.

“Claude Opus 5 one-shotted this game,” he wrote, adding that not a single external asset made it into the build.

Now, you may think such a good quality output required a long, detailed, and careful prompt to guide the AI models through the complexities of building a polished first-person shooter.

Think again.

The prompt behind it ran three short paragraphs, published in full on GitHub. It told Opus 5 to build a shooter at the level of the most recent Call of Duty games, to fan out subagents—workers that each get their own separate memory and a narrow job—to tackle pieces individually, and to keep looping on every piece with a separate, harsh critic until it held up against real Call of Duty footage in a blind side-by-side. The result, per the prompt, should be “utterly perfect.”

That’s a reversal of how prompt engineers have taught people to work. The advice through the vibe-coding boom was to specify criteria instead of adjectives: say what “good” means instead of just asking for it. What should the code consider instead of saying “AAA.”

Shumer’s version does close to the opposite, asking its own subagents to be “utterly wowed,” and leaving the actual definition to a critic Opus 5 built for itself.

That was close to the entire brief. Shumer later wrote that he never specified the renderer, listed the game’s systems, or defined what “AAA quality” needed to include. He has started calling the approach a Gauntlet Loop: hand an agent a real, inspectable bar instead of a vague instruction, let it split the job into small pieces, and route each piece through a critic that never sees the builder’s own reasoning for its choices.

Two Claude Code features carry that loop. Subagents spin up in isolated context windows with their own instructions and tool access, so a critic grading the weapon model doesn’t inherit the builder’s excuses for why it looks the way it does. Ultracode is a Claude Code setting that pushes the model to its top reasoning effort and lets it write its own orchestration plan, fanning work across as many as 16 agents at once, capped at 1,000 per run.

Anthropic’s built-in /loop skill, built for repeated fix-test-adjust cycles, is what kept the run from stopping the moment the game looked decent. Shumer never specified a number of rounds. He let the critic keep naming a new gap and kept the builder chasing it for hours before he closed the session himself.

The finished build runs on Three.js and plain WebGL2, with roughly 55,000 lines of code spread across 11 subsystems. Every texture, mesh, animation, and sound gets generated inside the browser at load time—no downloaded models, HDRIs, image files, or audio files. Shumer’s own published critic log shows the score climbing from 3.59 out of 10 toward just above 5, still trailing the real game every single round.

Skeptics assumed hours of hidden manual coding, so Shumer published the entire prompt and codebase. That’s when the copycats started.

Same trick, three different builders

James Altucher, the former hedge fund manager and podcaster, ran the identical prompt and reported spending “a little over ten hours” and roughly 1.3 million tokens on Opus 5 to get there. His build, Operation Blackout, plays free in the browser and looks awesome.

You can play that game here.

The developer of Prompt Silo, pointed the same request at OpenAI’s rival flagship instead, posting “Sol 5.6 Ultra with same prompt”—Sol being the top tier of the three-model GPT-5.6 family OpenAI made generally available July 9 alongside cheaper Terra and Luna versions.

Developer Leon Lin tried the opposite move, going for the usual detailed prompt. Rather than copy Shumer’s short version, he set out to “reverse engineer a prompt for this game,” producing a document running some 20 sections deep that spells out everything from ragdoll physics to cascaded shadow maps. He fed that into Cursor using plain Opus 5 on high effort, with no subagents and no ultracode, and the result—a market-street shooter called Dust Corridor—plays in-browser too, and also looks great.

None of the follow-up builds has faced the blind test Shumer ran on his own project. His critic log still shows real Call of Duty winning every round he logged—the bar Altucher and Atom Tan Studio are chasing with his exact three-paragraph prompt, and the one Leon Lin is chasing with roughly 20 sections of his own.

How much of this is actually new?

Agentic coding tools like Claude Code write software the way a supervised junior engineer might: They read files, run code, look at the screenshots they generate, and hand pieces of the job to subagents and critics that check the result against a stated goal. That loop is real, and Shumer’s Gauntlet Loop is a genuine way to structure it. None of that, on its own, proves the model designed a game from imagination rather than recombining code patterns it had already absorbed.

Three.js ships its own pointer-lock camera controls as an official example, and that base pattern—mouse-look, WASD movement, raycasting for gunfire—has been forked and tutorialized across GitHub, gists, and dev forums for more than a decade. A coding model trained on public repositories has almost certainly seen hundreds if not thousands of near-identical shooters before it ever read Shumer’s prompt.

That doesn’t make Claude of Duty fake, but it makes “built from scratch” a harder claim to fully credit, so take those results with a grain of salt.

Researchers who study code-generating models have a name for the broader issue: data contamination, when a model does well on a task mainly because near-identical examples already sat in its training data, not because it reasoned out something new.

None of the first-person shooter builds published a check for that kind of contamination. Shumer’s own repo does contain Claude’s own creativity, if it’s fair to call it that way. That’s a reason to read “one-shotted a AAA game” as a capable agent working inside one of the most heavily documented genres in programming, not as proof an AI designed a shooter with no prior art to lean on.