alepot55/agentrial
Statistical evaluation framework for AI agents
2 best points · 0 comments · 2 Hacker News posts · 20 GitHub stars · View on GitHub
Hacker News discussions
- Show HN: Agentrial – pytest for AI agents with statistical rigor (2 points, 0 comments, by alepot55, 2026-02-05)
- Show HN: Ran an AI agent 100x – pass rate 70%, not 100% (2 points, 0 comments, by alepot55, 2026-02-09)