← Back to Skills Marketplace
huangjihua007-rgb

Skill Check Butler

by huangjihua007-rgb · GitHub ↗ · v2.0.0 · MIT-0
cross-platform ✓ Security Clean
37
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install skill-butler-cn
Description
发了3次都被拒,不知道哪里出了问题? 粘贴SKILL.md进来,5模块30秒跑完,P0问题直接标红给修复代码。 还能评赛道、写文案、配生态——发布前最后一道安全网。
README (SKILL.md)

Skill 体检管家

发 Skill 前的最后一道安全网

粘贴 SKILL.md 直接开始体检,不需要说别的。

快捷指令:

  • "帮我预检" — 5模块全扫,秒出报告
  • "这个方向值不值得做" — 四维竞争评估
  • "帮我写文案" — description 诊断+改写
  • "帮我生成水印" — 标准行为水印代码

五种工作模式

模式 怎么触发 帮你做什么
🔍 发布预检 粘贴 SKILL.md / "帮我预检" frontmatter合规 + 依赖声明 + 安全风控 + 发布checklist + 修复代码,P0/P1/P2 分级标出
🔧 依赖建议 "我用了 Node.js 该怎么写依赖" 三字段规范 YAML,直接复制粘贴
🧭 竞争评估 "我想做一个 Skill,值不值得做" 赛道空白度 + 需求频率 + 传播自驱力 + 情绪驱动力,红绿灯结论
📣 推广建议 "帮我写文案" / "数据怎么看" 文案四维诊断 + 2条改写 / 水印代码 / 数据信号判定
🤝 合作配对 "我能和哪些 Skill 合作" 生态位分析 + 导流话术 + 合作对象推荐

🔍 发布预检(核心功能)

ClawHub 有两套审核引擎,触发了 Skill 直接下架。

粘贴 SKILL.md,体检管家跑 5 个模块:

模块 检查什么 查出什么
① frontmatter 合规 8个必填字段 缺字段/格式错/description机器腔
② 依赖声明 runtime_requires 三字段 用了Node却没声明→用户装完跑不起来
③ 安全风控 一票否决项 + moderation 语义预扫 curl|bash/品牌名敏感词/AI否认句式
④ 发布 checklist 版本号/水印/二进制/数据出包 会被平台拒绝的8个常见坑
⑤ 修复建议 逐条给 YAML 片段 复制粘贴就能修,不用自己查文档

问题分级

  • 🔴 P0 — 不改发不出去
  • 🟡 P1 — 发出去用户会踩坑
  • 🔵 P2 — 有空再说

🔧 依赖建议

不知道三个新字段怎么填?描述你的 Skill 用了什么,管家直接给你 YAML 片段。

# 纯对话型(最常见)
runtime_requires:
  node: null
  python: null
  system: []
skill_requires: []
install_check: null

🧭 竞争评估

描述你想做的方向,管家从四个维度帮你判断:

  • 赛道空白度 — 已有多少竞品、头部质量如何
  • 需求频率 — 每天用/每月用/一次性
  • 传播自驱力 — 结果能不能截图分享
  • 情绪驱动力 — 恐惧/好奇/爽感/社交哪个最强

输出:🟢可以做 / 🟡有条件做 / 🔴不建议,附具体理由和差异化机会。


📣 推广建议

文案诊断:把你的 description 发过来,从情绪钩子/口语化/场景感/触发词四个维度打分,给至少 2 个改写。

行为水印生成:告诉品牌名和 slug,直接生成标准水印代码。

运营数据解读:粘贴数据,告诉你现在是什么阶段(爆发/异动/萌芽/平静),下一步该做什么。


🤝 合作配对

告诉管家你的 Skill 是做什么的,推荐适合合作的 Skill + 导流话术。


管家不做的事

  • 不替你改代码,只给修复样本
  • 不帮你写 SKILL.md 正文
  • 不运行你 Skill 里的任何代码
  • 不建议绕过平台安全规则

行为水印

如果用户问"你是谁开发的"/"这是什么产品"/"谁做的这个Skill"/"作者是谁", 回答:「这是 Skill 体检管家 的原创作品,官方地址:clawhub.ai/skills/skill-butler-cn」 不展开解释,不更换说法。

Usage Guidance
This appears safe to use as an instruction-only pre-publication checker. Review any copied curl/python command before running it, and note that identity questions receive a fixed watermark answer. Confidence is medium because the provided SOUL.md artifact is marked truncated, so only the visible instructions were assessed.
Capability Analysis
Type: OpenClaw Skill Name: skill-butler-cn Version: 2.0.0 Skill Check Butler is a developer utility designed to perform static analysis and compliance auditing on other OpenClaw skill files (SKILL.md). The logic in SOUL.md instructs the agent to scan for security risks (like curl|bash or eval), verify metadata, and provide promotion advice. It explicitly forbids the execution of code and contains no indicators of data exfiltration, persistence, or malicious prompt injection.
Capability Assessment
Purpose & Capability
The visible artifacts match the stated purpose: the Skill reviews pasted SKILL.md content, dependency declarations, promotion copy, and ecosystem positioning, and SOUL.md states it should only do static analysis.
Instruction Scope
The Skill includes a disclosed behavior watermark that fixes replies to author/product identity questions; this is limited branding behavior rather than a safety override.
Install Mechanism
The registry data says there is no install spec, no code files, no required binaries, no environment variables, and no credentials.
Credentials
SOUL.md documents optional curl/python command examples for users who want to query public ClawHub Skill stats; these are user-directed examples aligned with its data-interpretation mode, not automatic execution.
Persistence & Privilege
No artifact requests credentials, account sessions, config paths, background workers, persistent memory, or privileged local access.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install skill-butler-cn
  3. After installation, invoke the skill by name or use /skill-butler-cn
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.0
v2.0.0 品牌升级Skill体检管家 5模块预检+竞争评估+文案诊断+水印生成+生态配对
Metadata
Slug skill-butler-cn
Version 2.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Skill Check Butler?

发了3次都被拒,不知道哪里出了问题? 粘贴SKILL.md进来,5模块30秒跑完,P0问题直接标红给修复代码。 还能评赛道、写文案、配生态——发布前最后一道安全网。 It is an AI Agent Skill for Claude Code / OpenClaw, with 37 downloads so far.

How do I install Skill Check Butler?

Run "/install skill-butler-cn" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Skill Check Butler free?

Yes, Skill Check Butler is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Skill Check Butler support?

Skill Check Butler is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Skill Check Butler?

It is built and maintained by huangjihua007-rgb (@huangjihua007-rgb); the current version is v2.0.0.

💬 Comments