---
name: gulmezeren2-byte/agent-evals
source: https://app.decimal.ai/s/gulmezeren2-byte-agent-evals@1/SKILL.md
source_sha256: 88c8cfbc0c32
---

# agent-evals

Measure an agent instead of vibing it: fixed task sets, seeded runs, and a
rubric written before the results are seen.