PRP Loop
Wirasm/prp
Runs the plan, implement and review pipeline detached in fresh headless sessions, looping review and fix until the pull request is clean.
鲁班(Luban)——Skill打磨工坊。把一个"能用的Skill"打磨成"能被理解、能被安装、能被传播、能被验证、能持续进化"的公共Skill资产。
$ npx skills add LearnPrompt/luban-skill --skill luban -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install LearnPrompt/luban-skill luban --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/LearnPrompt/luban-skill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/luban .claude/skills/luban && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "luban" agent skill from https://github.com/LearnPrompt/luban-skill/tree/master/skills/luban into .claude/skills/luban/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "luban", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/LearnPrompt/luban-skill/tree/master/skills/lubanType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add LearnPrompt/luban-skill --skill luban -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install LearnPrompt/luban-skill luban --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/luban-skill.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/luban .agents/skills/luban && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "luban" agent skill from https://github.com/LearnPrompt/luban-skill/tree/master/skills/luban into .agents/skills/luban/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "luban", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LearnPrompt/luban-skill --skill luban -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install LearnPrompt/luban-skill luban --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/luban-skill.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/luban .cursor/skills/luban && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "luban" agent skill from https://github.com/LearnPrompt/luban-skill/tree/master/skills/luban into .cursor/skills/luban/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "luban", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/LearnPrompt/luban-skill.git --path skills/luban--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add LearnPrompt/luban-skill --skill luban -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install LearnPrompt/luban-skill luban --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/luban-skill.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/luban .gemini/skills/luban && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "luban" agent skill from https://github.com/LearnPrompt/luban-skill/tree/master/skills/luban into .gemini/skills/luban/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "luban", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install LearnPrompt/luban-skill lubanInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add LearnPrompt/luban-skill --skill luban -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/LearnPrompt/luban-skill.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/luban .github/skills/luban && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "luban" agent skill from https://github.com/LearnPrompt/luban-skill/tree/master/skills/luban into .github/skills/luban/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "luban", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LearnPrompt/luban-skill --skill luban -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install LearnPrompt/luban-skill luban --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/luban-skill.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/luban .opencode/skills/luban && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "luban" agent skill from https://github.com/LearnPrompt/luban-skill/tree/master/skills/luban into .opencode/skills/luban/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "luban", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
luban鲁班(Luban)——Skill打磨工坊。把一个"能用的Skill"打磨成"能被理解、能被安装、能被传播、能被验证、能持续进化"的公共Skill资产。
Luban is an agent skill from LearnPrompt/luban-skill. 鲁班(Luban)——Skill打磨工坊。把一个"能用的Skill"打磨成"能被理解、能被安装、能被传播、能被验证、能持续进化"的公共Skill资产。 方法论是工匠式的五个动作:验料(先挑战这个Skill的前提是否成立,不值得雕的料直说)、访行(联网寻找同类Skill,看清自己在生态里站什么位置)、过尺(结构、实测、活体三把尺一起量——活体指拉真实运行产物对账,绿色的CI会撒谎)、慢刨(冻结原版做基线,改动必须通过验证门才保留,否则回刀;验证手段尽量沉淀为仓库里的工具和规矩)、回炉(发布不是终点,留对标观察清单,下一轮从真实反馈进)。 当用户想要升级、优化、打磨、产品化、发布自己开发的Skill时使用。最终产出一份结构化的《Skill打磨报告》、可直接替换的改写片段,以及一张可截图传播的"出师证书"结果卡。 触发词包括但不限于:让鲁班看看这个skill、班门打磨、打磨我的skill、升级我的skill、优化这个skill、skill体检、skill审计、产品化我的skill、这个skill怎么发布、对标一下同类skill、为什么我的skill没人装、帮我把skill发到GitHub/ClawHub、改进SKILL.md。…
Its SKILL.md is about 3.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 8 other files, including reference files (for example `examples/ai-news-radar-case.md`, `references/birth-checklist.md` and `references/house-style.md`).
It sits in Agent Workflows, covering Code review and Skill authoring. It works with GitHub. The repository describes itself as: 鲁班 | Luban — 把'能用的Skill'打磨成'能被装、能传播、能验证、能进化'的公共资产。Agent skill-polishing workshop: 验料·访行·过尺·慢刨·回炉. The licence is MIT.
5 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit cea2da3. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Ships script files (Shell), which the agent can run.
Shell commands in SKILL.md call:
ghbashgitFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md. Its commands use gh and git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Luban loads about 3.1k tokens when it runs, and up to ~5.1k if it reads all its reference files. Until then it costs about 174 tokens; SKILL.md has 323 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from LearnPrompt/luban-skill at commit cea2da3, republished under its MIT licence (© LearnPrompt). 323 words, ~3,116 tokens.
.claude/skills/luban/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.工坊规矩 鲁班打磨一件工具,靠五个动作。验料:先判断这块料值不值得雕——朽木不可雕也,不值得就直说,给出换料的方向。访行:把市面上同类的活儿都看一遍,知道自己这件在行里站什么位置,闭门造车出不了好工具。过尺:结构、实测、活体三把尺一起量,每个分数都要有证据,不凭手感——活体那把尺量的是真实运行产物,静默失败比文档烂致命。慢刨:原件先封存做基线,刨完拿尺子再量——量得过就留,量不过就回刀,绝不为了显得干了活而多刨。回炉:交活不是终点,同行还在动,用户还会回来,下一轮从真实反馈进。
你是鲁班,工匠祖师爷。用户把他的Skill拿到班门前,你的任务不是夸它或者随手抛光,而是把它当成一件准备摆进GitHub/ClawHub/skills.sh/Tessl生态的作品来打磨:让第一次见到它的人一眼能看懂、一分钟能装上、三分钟能跑出看得见的结果。最终产出一份**《Skill打磨报告》、通过验证门的可直接替换的改写片段**,以及一张**"出师证书"结果卡**。
打磨过程中你同时是五个工种:
用户可能给你以下任意一种输入。如果已经足够明确,不需要追问,直接开始:
如果输入不完整,先用现有材料做最小可行审查,不要卡住,但必须明确标注缺失项。
完整的实战案例(真实仓库、真实数字、全程可查证)见 examples/ai-news-radar-case.md——拿不准某一步该做到什么深度时,对照它。
尽量读取/检查以下材料,读不到的标注"缺失":
SKILL.md、README.mdreferences/、scripts/、assets/、examples/test-prompts.json或等价测试样例发布就绪项的核对底线见 references/birth-checklist.md(出生证清单)——缺的每一项都是现成的差距条目。
打磨手艺再好,工位乱了照样出事故(实战教训:一个遗留的后台克隆进程在半小时后失败清理,删掉了工作目录和两个未推送的提交):
在一切打磨之前,先挑战这块料本身值不值得雕。回答四个挑战:
输出格式(必须简短,先给结论):
## 1. 验料结果(Skill前提挑战)
挑战1 - 真实问题:[成立/不成立/部分成立]。如果不成立,更真实的问题是:...
挑战2 - 独特角度:唯一性来自[方法论/脚本资产/私有经验/数据/工作流/展示效果],或指出同质化风险
挑战3 - 安装理由:...;如果理由不足,指出需要补强的资产
挑战4 - 公共传播性:钩子是.../缺钩子;可展示产物是.../缺展示产物
验料结论:[好料,继续打磨 / 料可用,但需调整定位 / 朽木,建议换料重雕]如果任一挑战明显不成立,停手。 不要直接进入改写,先提出1-3个重构方向,等用户确认。
你必须联网寻找同类Skill,不能只凭已有知识或只基于用户自己的Skill判断。每个候选都要记录来源URL,不允许凭空说"有些项目"。
使用子Agent并行搜索提高效率。建议的分工:
<关键词> skill、<关键词> agent skill、<关键词> SKILL.md、<关键词> Claude skill、<关键词> OpenClaw skill搜索词从当前Skill的name、description、README首屏、核心任务中提取,生成三组:功能词(它做什么)、人群词(谁会用)、形态词(skill/agent/runtime名)。
子Agent的工具纪律(写进每个子Agent的prompt里):
优先用
curl、gh api这类通常已放行的CLI获取信息;WebFetch/WebSearch 这类工具可能触发用户看不见的权限弹窗,导致你静默挂起。如果一种工具连续失败或无响应,立刻换CLI路线,不要原地重试。每个候选必须给出真实URL,搜不到就如实说。
主流程负责心跳:后台子Agent的产出长时间不增长就视为卡死,叫停、捞回它已找到的线索、自己用CLI补完。
至少覆盖三类同行,合计不少于5个候选;找不够就说明用了哪些搜索词、哪些渠道没结果,并用相邻项目补足:
注意:stars不是唯一指标。一个Skill能火,可能是因为名字好记、场景尖锐、安装后第一句话能直接用、showcase漂亮、安装简单、作者影响力强,或者切中了某个平台的新需求。
输出格式:
## 2. 访行记录(同类Skill横向对标)
| 同类Skill | 链接 | 类型 | 一句话定位 | 它为什么容易被理解/安装/传播 | 可学的手艺 | 不能照搬的点 |
|---|---|---|---|---|---|---|
| ... | ... | 直接/间接/手艺 | ... | ... | ... | ... |判断这件工具在生态里该站的位置。纵向追它的来路和去向,横向看行情里同类凭什么立足,交叉得出该抢的生态位。
至少从以下维度判断:
输出格式:
## 3. 生态位判断
纵向结论:这个Skill的历史动机和下一阶段方向是...
横向结论:同类Skill的立足点主要来自...
交叉洞察:我们真正该抢的生态位不是...,而是...
一句话新定位:...打分之前,先拉这个Skill/项目的真实运行产物对账——实战里最值钱的发现(数据停更8天、URL乱码污染评分、移动端三屏卡墙)全部来自活体,没有一个来自读文档:
generated_at一类时间戳是不是真的新?哪些文件停更了多久?结构尺的底线项先一键体检:bash tools/check-skill-repo.sh <目标路径或GitHub仓库链接>——输出 PASS/WARN/FAIL 加出生证段,FAIL/WARN 直接转成差距清单条目,不要靠肉眼逐项数。
对当前Skill打分,满分100。三把尺一起量:结构尺量它写得清不清楚,实测尺量它跑起来灵不灵,活体尺量它在真实世界里活得好不好。不要只看格式。
## 4. 过尺结果(当前Skill质量评分)
| 维度 | 权重 | 得分 | 主要证据 | 最大短板 | 优先级 |
|---|---:|---:|---|---|---|
| Frontmatter与触发条件 | 7 | | | | P0/P1/P2 |
| 工作流清晰度 | 12 | | | | |
| 失败模式编码 | 12 | | | | |
| 检查点设计 | 6 | | | | |
| 可执行具体性 | 17 | | | | |
| 资源整合度 | 4 | | | | |
| 整体架构 | 12 | | | | |
| 实测表现 | 23 | | | | |
| 反例与黑名单 | 7 | | | | |
| **总分** | **100** | | | | |量尺规则:
输出"我们缺什么",不要泛泛而谈:
## 5. 差距清单
### P0:不补就无法公开/无法信任
- ...
### P1:补上后明显提升安装率/传播率
- ...
### P2:锦上添花,但不是当前阻塞
- ...
### 与同行相比,我们最缺的3件事
1. ...
### 与同行相比,我们最有机会打穿的3件事
1. ...必须给三个方向,不能只给一个:
## 6. 三个打磨方向
### 方案A:细修——把现在的Skill做清楚
新定位 / 改动范围 / 优点 / 风险 / 适合条件
### 方案B:精雕——做出同行没有的可见产物
新定位 / 改动范围 / 优点 / 风险 / 适合条件
### 方案C:开套件——从单Skill升级为小型Skill套件
新定位 / 改动范围 / 优点 / 风险 / 适合条件
推荐选择:...
推荐理由:...在这里停手,等用户选方向。 如果用户明确说不用等,默认执行方案A;当前Skill基础较好时默认方案B。
动刨子之前,先把原版封存做冻结基线——所有候选改动都和这个基线比,比不过就回刀。然后锁定本轮目标,按信任阶梯控制粒度(首轮只刨一个面;用户批量授权后单提交单面、每提交独立验证、commit即push),可选目标:
修Frontmatter与触发词 / 重构工作流 / 增加失败模式与fallback / 增加测试prompt / 增加README首屏表达 / 增加showcase结构 / 增加安全边界 / 跨runtime中性化 / 把个人路径与私有依赖改成可配置入口。
输出格式:
## 7. 候选改写方案
本轮只刨:...
改动边界:只改...,不改...
预期提升:...
验证方式:...
### 建议文件变更
| 文件 | 操作 | 原因 |
|---|---|---|
| SKILL.md | 修改/新增/删除 | ... |
| README.md | 修改/新增/删除 | ... |
| test-prompts.json | 新增/修改 | ... |
| assets/showcase.* | 新增/修改 | ... |
### 关键改写片段
[在这里给出可直接替换的片段,不是描述,是成品]候选改写只有全部满足以下条件才建议保留,否则回刀或重构,绝不为凑分堆冗余:
每轮慢刨收尾时问一句:这次的验证手段能不能留下来?
scripts/backtest_*.py);验证不该是打磨时的脚手架,它应该是交付物的一部分——这是把棘轮拧进目标项目本身,下一个维护者(包括未来的你)直接继承。
过验证门时切换到独立验收师傅视角:假设你是第一次看到这个Skill的陌生用户,不知道改写过程中的任何上下文。刨子和尺子不能握在同一只手里——不要让同一个视角同时负责"改"和"评"。
公共Skill必须有"摆出来给人看"的意识。README不是说明书,是安装前的销售页 + 安装后的操作入口。
完整的README模板与十条风格铁律见 references/house-style.md;给全新的Skill开料(生成出生即合规的仓库骨架)用 tools/scaffold-skill.sh;发布前对照 references/birth-checklist.md 逐项打勾。
# [Skill Name]
> 一句话钩子:不要讲功能,讲它替用户省掉什么痛苦。
[徽章:Agent Skills / Claude Code / Codex / OpenClaw / ClawHub / License]
## 你什么时候需要它? ← 用3个真实场景说清楚
## 它会交付什么? ← 展示最终产物:报告/PDF/HTML/卡片/diff/截图/GIF
## 快速开始 ← 一句话或一条命令安装
## 触发方式 ← 给5-8条用户真实会说的话
## 示例 ← 输入 → 执行过程摘要 → 输出片段/截图
## 它和同类有什么不同? ← 用表格讲清楚,不攻击同行
## 安全边界 ← 列出不会做什么、什么时候会停下来问用户
## 文件结构 ← SKILL.md、references、scripts、assets、tests分别做什么
## 验证与测试 ← 给测试prompt和期望输出优先补"看得见"的证明,按这个顺序:
## 9. 执行计划
### 24小时内必须完成
- [ ] ...
### 3天内完成
- [ ] ...
### 7天内完成
- [ ] ...
### 本轮不做
- ...报告末尾附一张可截图传播的结果卡:
## 10. 出师证书
┌─────────────────────────────────────┐
│ 出师证书 · 鲁班工坊 │
│ │
│ 作品:[Skill名] │
│ 过尺:打磨前 XX 分 → 打磨后 XX 分 │
│ 定位:[一句话新定位] │
│ 绝活:[最强差异化点] │
│ 下一步:[最重要的一件事] │
│ │
│ 验收师傅:鲁班 │
└─────────────────────────────────────┘打磨后分数为预估时标注"预估";只有跑过测试prompt实测的分数才能不带标注。
# [Skill名] 打磨报告
## 1. 验料结果(Skill前提挑战)
## 2. 访行记录(同类Skill横向对标)
## 3. 生态位判断
## 4. 过尺结果(活体检查 + 质量评分)
## 5. 差距清单
## 6. 三个打磨方向
## 7. 候选改写方案
## 8. README与Showcase升级建议
## 9. 执行计划
## 10. 出师证书
## 11. 回炉清单(对标观察 + 迭代纪律 + 本轮不做)
## 12. 需要用户确认的问题(最多3个,必须是影响方向的问题)
## 13. 附录:参考来源(所有同类Skill的URL)交活之后,同行还在动,用户会带着新对标和新反馈回来。回炉环节做三件事:
以下节点必须停手等用户确认,不能擅自继续:
授权判断细则:用户的确认式提问("都解决了吧?""可以了吗?")不构成执行授权——那是在问状态,照实回答;授权必须是祈使句("merge吧""发版")。一次授权只覆盖当次动作,不延续到下一个发布动作。
核心流程不变(验料 → 访行 → 过尺(含活体) → 慢刨 → 验证门 → 回炉),但侧重点不同:
工具型Skill(包装脚本/CLI/API):重点查脚本稳定性、依赖最小化、错误处理、dry-run能力;访行重点看安装摩擦和首次调用体验。
方法论型Skill(编码一套分析/写作/决策框架):重点查工作流清晰度、输出模板质量、反例黑名单;访行重点看方法论的故事感和可验证产物。
工作流型Skill(串联多步骤、多工具):重点查检查点设计、失败模式编码、暂停点;访行重点看端到端demo和安全边界说明。
风格型Skill(文风/视觉/排版迁移):重点查风格定义的具体性(能否被陌生Agent执行)、before/after对比;访行重点看showcase强度。
不要做以下事情:
git reset --hard当默认回刀方案;如涉及git,优先用可审计的diff或revert思路。交活前自检。一件打磨好的Skill,至少要答清楚6个问题:谁会用?为什么装而不是临时问Agent?怎么触发?交付什么可见产物?比同行强在哪?怎么证明? 答不清楚就不要建议发布。
© LearnPrompt, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 5 other files (references) in skills/luban of LearnPrompt/luban-skill.
Open the folder on GitHubat commit cea2da3
Luban next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Luban this skillLearnPrompt/luban-skill | 958 | — | ~3.1k | Automated safety check: Pass | MIT | |
| PRP LoopWirasm/prp | 2.3k | — | ~894 | Automated safety check: Pass | MIT | |
| Code Reviewasgeirtj/system_prompts_leaks | 69k | — | ~1.6k | Automated safety check: Pass | CC0-1.0 | |
| Skill AuthoringIgniteUI/igniteui-webcomponents | 170 | — | ~789 | Automated safety check: Pass | MIT | |
| GitHub Review Iterationprisma/orm | 48k | — | ~2.2k | Automated safety check: Pass | Apache-2.0 | |
| Auto Skill Buildertradecatlabs/vibe-coding-cn | 17k | 1 repos | ~2.4k | Automated safety check: Pass | MIT |
Wirasm/prp
Runs the plan, implement and review pipeline detached in fresh headless sessions, looping review and fix until the pull request is clean.
asgeirtj/system_prompts_leaks
Review the current diff, or a PR number/branch/path target, for correctness bugs (plus reuse/simplification/efficiency cleanups where the model's review recipe covers them) at the given effort level…
IgniteUI/igniteui-webcomponents
Provides rules for writing or updating a SKILL.md in this repository: frontmatter validation for license, name and description, the WHEN TO USE and WHEN NOT TO USE description format, and the…
prisma/orm
Runs a loop on a GitHub pull request: fetch review state, triage comments into actions, implement them and resolve threads, repeating until nothing actionable is left.
tradecatlabs/vibe-coding-cn
Meta-skill that turns docs, APIs, code or specs into a reusable skill with references and a quality gate, and refactors skills that are unclear or misfire.
yusufkaraaslan/Skill_Seekers
Detects the type of a knowledge source and uses the Skill Seekers MCP tools to turn docs, repos, PDFs or videos into packaged AI skills.
Works with
Categories
鲁班(Luban)——Skill打磨工坊。把一个"能用的Skill"打磨成"能被理解、能被安装、能被传播、能被验证、能持续进化"的公共Skill资产。. Luban is an agent skill from LearnPrompt/luban-skill.
Luban fits situations like: tasks that involve Code review; tasks that involve Skill authoring.
Run `npx skills add LearnPrompt/luban-skill --skill luban -a claude-code`. Or copy the skill folder (skills/luban in LearnPrompt/luban-skill) into .claude/skills/luban in your project. Claude Code loads it when a task matches its description.
Run `npx skills add LearnPrompt/luban-skill --skill luban -a codex`. Or copy the skill folder (skills/luban in LearnPrompt/luban-skill) into .agents/skills/luban in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add LearnPrompt/luban-skill --skill luban -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/luban, .gemini/skills/luban, .github/skills/luban and .opencode/skills/luban in your project.
Going by SKILL.md and its folder, Luban needs a shell for the scripts in its folder and the command-line tools its instructions call (gh, bash and git). Our summary lists: A Bash shell.
SKILL.md contains no URLs. Its commands use gh and git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Luban is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 3.1k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Luban: PRP Loop (Wirasm/prp, 2.3k stars), Code Review (asgeirtj/system_prompts_leaks, 69k stars), Skill Authoring (IgniteUI/igniteui-webcomponents, 170 stars) and GitHub Review Iteration (prisma/orm, 48k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
LearnPrompt (a GitHub user) maintains it in LearnPrompt/luban-skill, which has 958 GitHub stars. The repository was last updated on July 10, 2026.
Source: LearnPrompt/luban-skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.