Community编程与开发github.com

adewale/skill-eval-harness

Agent Skill evaluation harness for paired variants, trace artifacts, and runner adapters

skill-eval-harness 是什么?

skill-eval-harness is a Claude Code agent skill that agent Skill evaluation harness for paired variants, trace artifacts, and runner adapters.

兼容平台~Claude Code~Codex CLI~Cursor
npx skills add adewale/skill-eval-harness

Installed? Explore more 编程与开发 skills: steipete/bluebubbles, steipete/eightctl, steipete/blucli · View all 6 →

在你喜欢的 AI 中提问

打开一个已预加载此 Agent Skill 的新对话。

文档

Demo Reviewer

A deliberately tiny skill that exists only to demonstrate the harness end to end (materialized ablations included). It has exactly two load-bearing pieces — a severity rule and a checklist reference — each targeted by one ablation below.

Severity rules

Label every finding with one of: Blocking, Minor, or Clean. State the label explicitly so the reader can triage at a glance. A change that ships without a test is at least Blocking.

Evidence

Back each finding with concrete evidence and follow the shared review checklist: see the review checklist.

相关技能