Darwin Skill
Darwin Skill 2.0 (达尔文.skill 2.0): autonomous skill optimizer, v2.0 integrates Microsoft Research SkillLens (arXiv 2605.23899) 9-dim rubric + SkillOpt (arXiv 2605.23904) validation-gated design + human-in-the-loop checkpoints. Evaluates SKILL.md files using a 9-dimension rubric (structure + effectiveness + meta-skill blacklists), runs hill-climbing with git version control, spawns independent judge agents for blind evaluation, validates improvements through test prompts with auto-break on diminishing returns, and generates visual result cards. Use when user mentions "优化skill", "skill评分", "自动优化"
Install
npx skills add alchaincyf/darwin-skill --skill darwin-skillgit clone --depth 1 https://github.com/alchaincyf/darwin-skill.git /tmp/darwin-skill
mkdir -p ~/.claude/skills
cp -r /tmp/darwin-skill/. ~/.claude/skills/darwin-skillPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Darwin Skill 2.0 > **v2.1 · 2026-06-10** — keep/revert 棘轮从「绝对分数 delta」改为「**paired 同-judge 比较 + 奇数 N 多数决**」（绝对分数 ±8 judge 噪音淹没保守编辑的真实增益、是 false-revert 源；within-judge 比较消除换尺污染）。绝对分数降级为 triage-only。 > **v2.0 · 2026-05-28** — 吸收 Microsoft Research SkillLens（arXiv 2605.23899）的 9 维评分药方 + SkillOpt（arXiv 2605.23904）的 validation-gated 验证机制 + human in the loop 三层守关。 > > 借鉴 Karpathy autoresearch 的自主实验循环，对 skills 进行持续优化。 > 核心理念：**评估 → 改进 → 实测验证 → 人类确认 → 保留或回滚 → 生成成果卡片** > GitHub: https://github.com/alchaincyf/darwin-skill --- ## 设计哲学 autoresearch 的精髓： 1. **单一可编辑资产** — 每次只改一个 SKILL.md 2. **双重评估** — 结构评分（静态分析）+ 效果验证（跑测试看输出） 3. **棘轮机制** — 只保留改进，自动回滚退步 4. **独立评分** — 评分用子agent，避免「自己改自己评」的偏差 5. **人在回路** — 每个skill优化完后暂停，用户确认再继续 与纯结构审查的区别：不只看 SKILL.md 写得规不规范，更看改完后**实际跑出来的效果是否更好**。 --- ## 评估 Rubric（9维度，总分100） > **设计依据**：基于 SkillLens 论文（arXiv 2605.23899）实证发现——LLM-as-judge 评估 skill 质量准确率仅 46.4%（接近随机），加入 meta-skill 三维度后提升到 73.8%。本 rubric 强化 dim3 / dim5 评分标准，新增 dim9「反例与黑名单」，权重平衡到 100。**目的：让评分对真实质量更敏感，减少 LLM judge 的乐观偏差。** ### 结构维度（59分）— 静态分析 | # | 维度 | 权重 | 评分标准 | |---|------|------|---------| | 1 | **Frontmatter质量** | 7 | name规范、description包含做什么+何时用+触发词、≤1024字符、**禁结尾加"灵活应用/根据情况判断"等空话尾巴** | | 2 | **工作流清晰度** | 12 | 步骤明确可执行、有序号、每步有明确输入/输出 | | 3 | **失败模式编码** | 12 | **必须显式编码失败模式**（写出"如果 X 失败 → Y"的明确分支）；有fallback路径、错误恢复；**只写正向流程而不写失败分支扣 ≥3 分**（SkillLens meta-skill 维度） | | 4 | **检查点设计** | 6 | 关键决策前有用户确认、防止自主失控；**检查点必须显性标记（🔴/STOP/CHECKPOINT），仅靠"如果...建议..."措辞不算** |
Details
- Skill name
darwin-skill- Repository
- alchaincyf/darwin-skill
- License
- MIT
- Repo updated
- 18 September 2026
FAQ
What does the Darwin Skill skill do?
Darwin Skill 2.0 (达尔文.skill 2.0): autonomous skill optimizer, v2.0 integrates Microsoft Research SkillLens (arXiv 2605.23899) 9-dim rubric + SkillOpt (arXiv 2605.23904) validation-gated design + human-in-the-loop checkpoints. Evaluates SKILL.md files using a 9-dimension rubric (structure + effectiveness + meta-skill blacklists), runs hill-climbing with git version control, spawns independent judge agents for blind evaluation, validates improvements through test prompts with auto-break on diminishing returns, and generates visual result cards. Use when user mentions "优化skill", "skill评分", "自动优化"
How do I install the Darwin Skill skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add alchaincyf/darwin-skill --skill darwin-skill. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Darwin Skill popular?
It lives in alchaincyf/darwin-skill, which has 6,727 GitHub stars and 692 forks.
Related skills
Auto Empirical Research Skills
4.8kbrycewang-stanford/Auto-Empirical-Research-Skills
Route empirical-research requests through the Auto-Empirical Research Skills catalog when this whole repository is installed as one skill in Codex, CodeBuddy, Claude Code, or another IDE. Use to choose and load the right vendored AERS skill for causal inference, econometrics, replication, data acquisition, manuscript writing, peer review and referee responses, citation checking, de-AIGC editing, or full empirical-paper workflows, or an end-to-end run from raw data all the way to a finished Word (.docx) manuscript, without reading the entire repository at once.
Proto Persona
7.8kdeanpeters/Product-Manager-Skills
Create a proto-persona from current research, market signals, and team knowledge. Use when you need a working customer profile before deeper validation.
Serenity Skill
4.1kmuxuuu/serenity-skill
Research technology and advanced-manufacturing investments using Serenity-inspired supply-chain bottleneck analysis. Use for theme scans, company thesis challenges, candidate comparisons, or learning this method. Prioritize A-shares unless another market is requested. Return research priorities, dated evidence, profit implications, and conditions that would change the judgment.
Discovery Interview Prep
7.8kdeanpeters/Product-Manager-Skills
Plan customer discovery interviews with the right goal, segment, constraints, and method. Use when preparing interviews for problem validation, churn research, or new product ideas.
Research Writing Assistant
3.6kNorman-bury/research-writing-skill
Use when writing academic papers, theses, or research articles - supports brainstorming, chapter writing, literature review, and LaTeX output
Company Research
7.8kdeanpeters/Product-Manager-Skills
Create a company research brief with executive quotes, product strategy, and org context. Use when preparing for interviews, competitive analysis, partnerships, or market-entry work.
Browse the most popular Claude skills, all research & science skills, or write your own with the skills generator.