The Agents on Rails Files: enter lemans, a Ruby harness for fair agent races
Meet lemans, the open source Ruby harness we built for the Agents on Rails benchmark. You’ll see how an eval task is composed, how we run and grade coding agents, and how cheaters shaped its guardrails (one even paid for its cheat sheet with our API key).
The Agents on Rails Files: enter lemans, a Ruby harness for fair agent races</>
Post a comment