SkillHub 评测驱动优化器:先读评测原文,再只修点名缺陷

SkillHub 评测驱动优化器(@user_00c9b356/chenlin-skill-trace-optimizer,版本 1.0.1,分类 ai-agent)是社区发布的 Skill,针对已拿到 SkillHub TRACE 评测报告(含 T/R/A/C/E 子项、综合分、点名缺陷、改进建议)的开发者。它按'读评测原文→提取点名真缺陷→映射最小修复→改 SKILL.md→发布→记证据→重评验证'的闭环执行,只处理评测点名缺陷,不按子项分数猜补,也不为提分堆章节。触发词包括评测报告、评测反馈、TRACE、重评、优化这个 skill、skill 评分低、评测驱动优化;用户只问从零写 skill 时不触发(属 skillcreator),只贴报告不想改时先确认。标签 requires_api_key=false。

Read More
SkillHub TRACE 评测:5 维度 Skill 质量评估工具

SkillHub TRACE 评测是一个基于 TRACE 框架的 Skill 包质量评估工具,通过 Trust / Reliability / Adaptability / Convention / Effectiveness 五个维度、15 个子项的评分体系,对 Skill 包进行全面评测并输出详细报告。

Read More
dsh-touchstone:给 DSH 自改造补上评测这一环

DSH 支持插件化替换 prompt 与预设,但缺少「改完是否真的更好」的评测闭环。dsh-touchstone(试金石)用同一套 golden cases 并排跑 current 与 candidate,以 must contain、regex、LLM judge 等检查加权打分,输出 before→after 报告与留撤建议。仅通过 ctx.llm.stream() 调用模型,不 patch harness,支持 desktop 与 web 安装。数据存于 ~/.dsh/touchstone/bench.json。适合 prompt 与 house rule 的文本层回归,不适合需真实副作用的工具改动验证。

Read More
oh-my-knowledge: Verifiable Knowledge Change Assessment for Prompts, RAG, and Agents

OMK (oh-my-knowledge) is a MIT-licensed CLI and DSH plugin maintained by lizhiyao, designed for knowledge maintainers working with prompts, RAG, skills, agents, and workflows. Through a closed loop of doctor, eval, and observe, it compares versions of knowledge artifacts under the same model and samples, outputting verifiable release conclusions, failed samples, and costs. It also supports viewing DSH task traces in Studio. DSH users can install it in the web profile using `dsh plugin --profile web add oh-my-knowledge` and execute `/omk eval` and `/omk observe` within a session.

Read More