← All skills← 全部 skills

kelly-agent-eval

Review board that runs a fixed suite of mock test cases against baseline vs candidate agent versions, surfacing rubric-scored regressions before a release gate.评测看板，跑固定测试集对比 baseline 与候选版本 agent，在发版前用打分规则揪出回归问题。

Install安装 Install kelly-agent-eval or copy a ready prompt for your agent.安装 kelly-agent-eval，或复制现成提示给 agent。

npx skills add mr-kelly/skills
--skill kelly-agent-eval

Install kelly-agent-eval from https://mr-kelly.github.io/skills/ for this agent, then use it for my next task.

Run:
npx skills add mr-kelly/skills --skill kelly-agent-eval

After it is installed, invoke the skill as $kelly-agent-eval.

Claude CodeClaude Code

/plugin marketplace add mr-kelly/skills
/plugin install mr-kelly-skills

When to use it什么时候用

Use it when triaging agent-version regressions, comparing baseline vs candidate quality, or recording a release approve/block decision.排查 agent 版本回归、对比 baseline 与候选质量，或记录发版批准/阻止决定时使用。

**Overview**Baseline vs candidate pass-rate comparison and release approve/block panel.
**Overview**Baseline vs candidate pass-rate comparison and release approve/block panel.

**Case detail**Rubric bars and a side-by-side transcript diff for a regression.
**Case detail**Rubric bars and a side-by-side transcript diff for a regression.