Claude decided on its own to build pixel-by-pixel tests for Quake
Working on a Quake project, Claude decided on its own to build a test suite: demo playbacks with snapshots compared pixel by pixel, plus recorded demos whose screenshots it checked for each element.
Done withClaudeSpotted· by HN user magicalhippo· Oct 22, 2026
Result Used at most 1% of a $20 plan's weekly limit
“So now I have a solid set of tests to verify against.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
What they did
They were working on a Quake port when Claude, on its own initiative, decided to build demo playbacks and take periodic snapshots to compare pixel by pixel for differences. It saw the original Quake code hadn’t been modified, so it compiled a native version itself to use for the comparison, applying a small patch it found on its own to make the game deterministic.
How it went
Claude asked them to record a demo containing specific elements, like an explosion or being underwater, then visually verified the screenshots showed those elements. The result was a solid set of tests to verify against, using at most 1% of the weekly limit on their $20 plan.
Try it yourself with Claude
Build a visual regression test suite for [my game/project]: run demo recordings, take periodic screenshots, and compare them pixel by pixel to catch differences. If determinism is needed, find and apply whatever patch is required so repeated runs produce identical output, then verify the screenshots show the expected in-game elements.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.
After $45 in GitHub Actions overage in one month, a solo founder put an idle homelab to work as CI runners, with Claude making it easy to set up. Daily spend dropped from $2-4 a day to under $0.25.
A first homemade synth has 36 hall-effect keys and buttons and 7 rotary encoders. The firmware started by hand, then gentle prodding at Claude got everything needed much faster. Rev 2 of the PCB is nearly done.
After building a minimal proof assistant, the writer asked Claude how they knew the prover was actually sound. Claude checked, found five different ways it could prove false theorems, and fixed them.
Four years after shelving a homemade RSS reader and selling off its $500 domain, the writer built it this year with Codex and Claude. They use it daily, along with 30 other strangers who found it public.
Chasing a childhood itch for the exact 1996 After Dark Flying Toasters screensaver, the writer turned to Claude. It did a lot of good reverse-engineering work, but kept inventing details that needed pushback.
A rabbitmq connector occasionally got stuck after rare network drops, too minor to spend days fixing. Claude built a test harness that reproduced it, and the actual fix wasn't too hard.