Eval Harness

Use when setting up eval-driven development (EDD) for AI-assisted workflows, defining pass/fail criteria for agent task completion, measuring reliability with pass@k metrics, creating regression suites for prompt or agent changes, or benchmarking performance across model versions.

oyi77 Updated 19 repo stars

File contents

oyi77/1ai-skills commit 0be3624a85

Frequently asked questions

npx get-skill add oyi77/eval-harness