Community程式設計與開發github.com

helloJamest/SkillBench

SkillBench is an open-source framework for evaluating and evolving Codex skills with eval-case generation, judge-only/full-agent scoring, traceable dashboards, CI gates, and GEPA-style feedback loops for improving skill docs.

SkillBench 是什麼?

SkillBench is a Codex agent skill that skillBench is an open-source framework for evaluating and evolving Codex skills with eval-case generation, judge-only/full-agent scoring, traceable dashboards, CI gates, and GEPA-style feedback loops for improving skill docs.

相容平台~Claude CodeCodex CLI~Cursor
npx skills add helloJamest/SkillBench

Installed? Explore more 程式設計與開發 skills: steipete/bluebubbles, steipete/eightctl, steipete/blucli · View all 6 →

在你喜歡的 AI 中提問

開啟一個已預先載入此 Agent Skill 的新對話。

說明文件

SkillBench 是做什麼的?

SkillBench is an open-source framework for evaluating and evolving Codex skills with eval-case generation, judge-only/full-agent scoring, traceable dashboards, CI gates, and GEPA-style feedback loops for improving skill docs.

相關技能