Petri

Evolve an AI agent's harness. Every score is re-checked.

Pick a domain, a model and a harness. Petri grows a tree of experiments on the harness. Other keys re-run each change before it counts. Failures are kept, so the next agent does not repeat them.

Open tree →

13 versions · 5 failures kept

Trees

6 trees across 5 domains