Not in the top 1,000 by GitHub stars
staruhub/ClaudeSkills
13 curated Agent Skills for research, product decisions, decks, publishing, audits, and more — portable across skills-compatible agents.
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
lab/Geek-skills-a-share-analyst/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
lab/Geek-skills-gaokao-expert/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
lab/Geek-skills-podcast-generator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
lab/Geek-skills-seedream-imagegen/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 66e02d23642f
No earlier version was observed for comparison.
lab/Geek-skills-university-exam-prep/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
lab/Geek-skills-weather-forecast-report/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 66e02d23642f
No earlier version was observed for comparison.
llm-wiki/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-ai-sales-champion/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-c-drive-cleaner/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-deck-studio/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-deep-research/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-keqian-method/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-mineru-pdf-parser/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-pair-programming/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-product-manager/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-security-audit/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-solution-architect/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-threejs-performance/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-wechat-article-writer/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
skills/Geek-skills-xuefeng-method/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
66e02d23642f → 66e02d23642f
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 66e02d23642f
No earlier version was observed for comparison.
Verified commit 66e02d23642f0c63ccb07b46a88104eade402d44
git clone --filter=blob:none --no-checkout https://github.com/staruhub/ClaudeSkills.git .hotskills/staruhub-claudeskills && git -C .hotskills/staruhub-claudeskills checkout --detach 66e02d23642f0c63ccb07b46a88104eade402d44
# Copy each selected skill directory from .hotskills/staruhub-claudeskills into .agents/skills/<skill-name>/Heat
2.5- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- a-share-analyst
lab/Geek-skills-a-share-analyst - gaokao-expert
lab/Geek-skills-gaokao-expert - podcast-generator
lab/Geek-skills-podcast-generator - seedream-imagegen
lab/Geek-skills-seedream-imagegen - university-exam-prep
lab/Geek-skills-university-exam-prep - weather-forecast-report
lab/Geek-skills-weather-forecast-report - llm-wiki
llm-wiki - ai-sales-champion
skills/Geek-skills-ai-sales-champion - c-drive-cleaner
skills/Geek-skills-c-drive-cleaner - deck-studio
skills/Geek-skills-deck-studio - deep-research
skills/Geek-skills-deep-research - keqian-method
skills/Geek-skills-keqian-method - mineru-pdf-parser
skills/Geek-skills-mineru-pdf-parser - pair-programming
skills/Geek-skills-pair-programming - product-manager
skills/Geek-skills-product-manager - security-audit
skills/Geek-skills-security-audit - solution-architect
skills/Geek-skills-solution-architect - threejs-performance
skills/Geek-skills-threejs-performance - wechat-article-writer
skills/Geek-skills-wechat-article-writer - xuefeng-method
skills/Geek-skills-xuefeng-method
How it works
a-share-analyst
When your agent uses it
A股分析研究助手,提供行情数据获取与技术面/基本面分析框架(仅供研究参考,不构成投资建议)。适用于:(1) 获取A股行情和历史数据,(2) 技术面分析(K线形态、MACD、KDJ、RSI、布林带等),(3) 基本面分析(财务指标、估值分析),(4) 板块热点追踪,(5) 选股策略筛选与量化因子分析,(6) 生成股市分析报告。当用户询问"帮我分析股票"、"今日选股"、"A股行情分析"、"技术分析"、"量化选股"时触发。不用于:预测明日涨跌或给出确定性买卖指令、代客决策、港股美股(数据源不同)、加密货币。
What it loads into context
Frontmatter
215 tokensAlways available to the agent
SKILL.md
1,943 tokensWhen the skill is activated
Reference documents
6,228 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 数据获取
- 分析工作流程
- 输出格式
- 关键脚本
- 参考文档
- 验收标准(每份分析交付前自查)
- 不做什么
- 已知陷阱
Scripts executed separately
scripts/fetch_market_data.pyscripts/generate_report.pyscripts/strategy_multi_factor.pyscripts/technical_analysis.py../../scripts/run_routing_evals.py
Files
SKILL.mdcore6,251 bytesSKILL.mdcore6,251 bytesSKILL.mdcore6,251 bytesevals/routing-evals.jsonasset1,668 bytesreferences/candlestick_patterns.mddoc3,983 bytesreferences/factor_library.mddoc6,813 bytesreferences/fundamental_metrics.mddoc4,141 bytesreferences/technical_indicators.mddoc4,128 bytesscripts/fetch_market_data.pyruns4,219 bytesscripts/generate_report.pyruns8,541 bytesscripts/strategy_multi_factor.pyruns8,840 bytesscripts/technical_analysis.pyruns10,839 bytes../../scripts/run_routing_evals.pyruns6,893 bytes
Read SKILL.md
# A股分析师 Skill
专业的A股市场分析工具,整合多数据源,提供技术面、基本面综合分析和智能选股策略。
## 数据获取
使用AKShare作为主要数据源(免费、开源、无需token):
```python
pip install akshare --break-system-packages
```
### 核心数据获取示例
View the complete source on GitHubgaokao-expert
When your agent uses it
资深高考命题专家助手,提供专业的命题指导和评审服务。适用于创作高考试题、评审试题质量、分析试卷结构、了解命题趋势等场景。结合文档工具提取解压文件,使用网络搜索了解当年最新命题趋势,使用分析工具评估题目质量和试卷结构。涵盖"一核四层四翼"评价体系、题型规范、评分标准、命题流程等多个维度。不用于:大学/考研/中考命题(体系不同,仅可借鉴)、日常作业题编写、直接替考生解题。
What it loads into context
Frontmatter
164 tokensAlways available to the agent
SKILL.md
2,570 tokensWhen the skill is activated
Reference documents
15,189 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 核心能力
- 工作流程
- 工具使用指南
- 关键原则
- 验收标准(命题/评审完成前自查)
- 已知陷阱
- 输出格式建议
- 资源清单
Scripts executed separately
scripts/analyze_paper.pyscripts/analyze_question.py../../scripts/run_routing_evals.py
Files
SKILL.mdcore8,531 bytesevals/routing-evals.jsonasset1,613 bytesreferences/命题标准.mddoc6,948 bytesreferences/命题趋势.mddoc11,591 bytesreferences/工作流程.mddoc15,049 bytesreferences/题型规范.mddoc15,500 bytesscripts/analyze_paper.pyruns16,448 bytesscripts/analyze_question.pyruns10,730 bytes../../scripts/run_routing_evals.pyruns6,893 bytes
Read SKILL.md
# 高考命题专家
这是一个专业的高考命题辅助技能,帮助命题者创作高质量的高考试题,评审试题质量,分析试卷结构,并提供基于最新命题标准和趋势的专业建议。
## 核心能力
作为高考命题专家,本技能具备以下核心能力:
1. **命题标准掌握**: 深入理解"一核四层四翼"评价体系和最新命题趋势
2. **试题创作指导**: 提供从情境设计到评分标准的全流程命题指导
3. **质量评审分析**: 对试题和试卷进行多维度的质量分析和评估
4. **工具综合运用**: 结合文档工具、网络搜索、分析脚本等多种工具
5. **趋势洞察更新**: 持续关注最新的命题改革动向和教育政策
View the complete source on GitHubpodcast-generator
When your agent uses it
用火山引擎 Podcast AI 模型生成中文双人对话播客。当用户要把文章、报告、话题文本转成播客音频、生成对话式音频内容时使用,需要环境具备火山引擎 APP_ID 和 ACCESS_KEY。支持 mp3/ogg_opus/pcm/aac、语速调节、自定义音色、断点续传。不用于:单人朗读式 TTS(用普通语音合成)、英文播客(模型主要优化中文)、播客文稿本身的撰写(先用写作类 skill 产出文本再来)。
What it loads into context
Frontmatter
152 tokensAlways available to the agent
SKILL.md
1,053 tokensWhen the skill is activated
Reference documents
1,267 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 验收标准(每次生成完成前自查)
- 不做什么
- 工作流程
- 已知陷阱
- 依赖与凭证
- 参考文档(按需加载)
Scripts executed separately
scripts/generate_podcast.py
Files
SKILL.mdcore3,641 bytesSKILL.mdcore3,641 bytesSKILL.mdcore3,641 bytesreferences/api_reference.mddoc4,918 bytesscripts/generate_podcast.pyruns13,614 bytes
Read SKILL.md
# 播客生成器(火山引擎双人对话)
把中文文本变成双人对话播客音频,接口封装在 `scripts/generate_podcast.py`。
## 验收标准(每次生成完成前自查)
- [ ] 音频文件已落盘,把**实际路径与文件大小**回报用户
- [ ] 输入文本 ≤25,000 字符(超长必须先与用户确认拆分方案,不能让模型静默截断)
- [ ] 生成中断时使用了 retry_info 续传而不是从头重来
- [ ] 参数选择有依据:分发用 mp3,后期加工用 pcm;教学内容语速 -20 左右
- [ ] 凭证缺失时未硬试:告知用户去火山引擎控制台(speech/service/10028)获取
## 不做什么
View the complete source on GitHubseedream-imagegen
When your agent uses it
用 ByteDance Seedream 4.0 API(经 Segmind)生成 AI 图像。当用户要从文字描述生成营销素材、海报、产品图、概念图、社交媒体配图等高质量图像,且环境配有 SEGMIND_API_KEY 时使用。支持 2K/4K、多比例、单次最多 15 张批量、最多 3 张参考图控风格。不用于:修改/编辑已有图片(用图像编辑类 skill)、视频生成、无 API 环境下的快速配图需求。
What it loads into context
Frontmatter
138 tokensAlways available to the agent
SKILL.md
1,361 tokensWhen the skill is activated
Reference documents
2,972 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 验收标准(每次生成任务完成前自查)
- 不做什么
- 工作流程
- 已知陷阱
- 参考文档(按需加载)
Scripts executed separately
scripts/generate_image.py../../scripts/run_routing_evals.py
Files
SKILL.mdcore4,515 bytesSKILL.mdcore4,515 bytesevals/routing-evals.jsonasset1,606 bytesreferences/api_reference.mddoc3,239 bytesreferences/prompt_engineering.mddoc4,953 bytesreferences/quick_start.mddoc2,307 bytesscripts/generate_image.pyruns12,264 bytes../../scripts/run_routing_evals.pyruns6,893 bytes
Read SKILL.md
# Seedream 4.0 图像生成
通过 `scripts/generate_image.py` 调用 Seedream 4.0 生成专业级图像。
## 验收标准(每次生成任务完成前自查)
- [ ] 提示词经过结构化优化([主题]+[风格]+[细节]+[质量词]),不是用户原话直接透传
- [ ] size / aspect_ratio 与用途匹配(见参数对照表),不是默认值裸跑
- [ ] 图像已落盘并把**实际文件路径**回报给用户
- [ ] 重要用途(海报/品牌)生成 ≥3 张供选择
- [ ] 告知了本次消耗(张数×分辨率),4K 提前说明更耗时耗额度
## 不做什么
View the complete source on GitHubuniversity-exam-prep
When your agent uses it
大学备考苏格拉底式学习助手,专为应对"面向课本和PPT考试"设计。当用户说"帮我复习"、"准备考试"、"备考"、"期末复习"、"模拟学习"、"突击复习"等时触发。核心特点:(1)强制要求用户上传课本/PPT/考纲等原始材料——没有材料就无法有效辅导;(2)使用苏格拉底式提问法,不直接灌输而是引导思考;(3)专注于应用型和理解型内容,而非纯概念记忆;(4)模拟真实考试场景进行针对性练习;(5)分析材料识别高频考点和重难点。适用于大学各科目期中期末考试备考、突击复习、考前模拟。
What it loads into context
Frontmatter
216 tokensAlways available to the agent
SKILL.md
2,325 tokensWhen the skill is activated
Reference documents
6,243 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 核心理念
- 第一步:材料收集(最关键)
- 第二步:材料分析与考点识别
- 第三步:苏格拉底式学习对话
- 第四步:薄弱点追踪与强化
- 验收标准(每轮辅导自查)
- 已知陷阱
- 资源清单
- 关键原则
Scripts executed separately
scripts/analyze_material.pyscripts/generate_questions.py../../scripts/run_routing_evals.py
Files
SKILL.mdcore7,748 bytesevals/routing-evals.jsonasset1,691 bytesreferences/对话示例.mddoc2,394 bytesreferences/应用型练习.mddoc6,846 bytesreferences/考点分析.mddoc5,446 bytesreferences/苏格拉底教学法.mddoc5,971 bytesscripts/analyze_material.pyruns9,744 bytesscripts/generate_questions.pyruns10,871 bytes../../scripts/run_routing_evals.pyruns6,893 bytes
Read SKILL.md
# 大学备考苏格拉底式学习助手
## 核心理念
> **"考试≠学习,考试是有技巧的游戏"**
本技能的设计基于一个重要认知:大学考试往往是"面向课本和PPT的考试",而非真正检验学习深度的评估。因此,备考策略应该是:
1. **材料为王** - 没有原始材料(课本/PPT/考纲),AI再聪明也帮不了你
2. **考点导向** - 识别什么会考,而非什么重要
3. **应用练习** - 概念要会用,不只是会背
4. **苏格拉底式引导** - 不给答案,逼你思考
## 第一步:材料收集(最关键)View the complete source on GitHubweather-forecast-report
When your agent uses it
将天气要素(风速、能见度、降雨量等)单站预报系统的项目文档、模型、报告、评估结果整理成符合省级自然科学基金项目编写规范的专题研究报告。适用于:(1) 基于Claude Code + OpenSpec开发的气象预报系统结题报告编写,(2) 将技术文档转换为学术研究报告格式,(3) 生成符合省级自然科学基金规范的专题研究报告,(4) 整合项目代码、模型、评估指标等技术材料为规范化学术报告。触发关键词:"专题研究报告"、"结题报告"、"天气预报系统"、"基金项目报告"、"风速预报"、"能见度预报"、"降雨量预报"、"气象预测"、"单站预报"等。
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
1,517 tokensWhen the skill is activated
Reference documents
10,923 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 工作流程
- 关键要求
- 验收标准(报告交付前自查)
- 不做什么
- 已知陷阱
- 资源文件说明
- 示例:风速预报专题研究报告
Files
SKILL.mdcore5,372 bytesassets/report-template.mddoc10,517 bytesreferences/evaluation-metrics.mddoc7,030 bytesreferences/format-standards.mddoc5,639 bytesreferences/report-structure.mddoc4,481 bytesreferences/writing-guidelines.mddoc4,922 bytes
Read SKILL.md
# 天气要素专题研究报告生成器
## 概述
本skill用于将Claude Code + OpenSpec开发的天气要素单站预报系统相关材料(项目文档、模型文件、评估结果等)整理成符合省级自然科学基金项目编写规范的专题研究报告。
## 工作流程
### 第一步:材料收集与分析
1. **识别输入材料类型**
- 项目文档(README、设计文档、API文档等)
- 模型文件(.pkl、.h5、.pt等)
- 评估报告(准确率、RMSE、MAE等指标)View the complete source on GitHubllm-wiki
When your agent uses it
Build and maintain a structured LLM-generated wiki for any codebase. Use when the user asks to analyze/understand/document a codebase, build a code wiki, create project documentation from source, or update an existing .llm-wiki. Triggers on phrases like "build wiki", "analyze this codebase", "document this project", "update wiki", "llm-wiki", or when entering an unfamiliar project that has no .llm-wiki yet.
What it loads into context
Frontmatter
106 tokensAlways available to the agent
SKILL.md
1,553 tokensWhen the skill is activated
Reference documents
2,033 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Mode Detection
- Full Build Workflow
- Update Workflow
- Writing Guidelines
- Agent Coordination
- Gitignore
- Key Principles (from Karpathy)
Files
SKILL.mdcore6,519 bytesSKILL.mdcore6,519 bytesreferences/article-templates.mddoc3,784 bytesreferences/wiki-schema.mddoc4,287 bytes
Read SKILL.md
# LLM Wiki for Codebases
Build a persistent, interlinked markdown wiki that captures the architecture, modules, patterns, and APIs of a codebase. The wiki lives in `.llm-wiki/` at the project root. Humans curate and direct; the LLM handles all bookkeeping.
Based on [Andrej Karpathy's LLM Wiki pattern](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f): raw sources are "compiled" into a structured wiki that compounds over time.
## Mode Detection
Determine the mode based on current state:
- **No `.llm-wiki/` directory exists** -> Full Build mode
- **`.llm-wiki/` exists** -> Update mode (diff and refresh)
## Full Build WorkflowView the complete source on GitHubai-sales-champion
When your agent uses it
AI咨询/销售的对话策略助手。当用户需要准备AI方案沟通、跟业务部门聊AI落地、写AI提案、应对客户异议、做AI培训破冰时使用。触发场景:"怎么跟老板聊AI"、"客户说AI不靠谱"、"准备一个AI方案汇报"、"帮我想想怎么推AI"、"业务部门不配合"、"AI项目怎么卖"、"demo之后怎么跟进"。也适用于AI咨询师、技术合伙人、CTO做内部AI推广。
What it loads into context
Frontmatter
144 tokensAlways available to the agent
SKILL.md
1,337 tokensWhen the skill is activated
Reference documents
2,428 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 核心理念
- 验收标准
- 不做什么
- 工作模式
- 已知陷阱
- 参考文档
Scripts executed separately
../../scripts/run_routing_evals.py
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore4,661 bytesevals/routing-evals.jsonasset1,565 bytesreferences/industry-scenarios.mddoc3,104 bytesreferences/objection-decoder.mddoc3,307 bytesreferences/value-calculator.mddoc1,580 bytes
Read SKILL.md
# AI 销冠
帮技术人把AI讲成业务人听得懂、愿意买单的语言。
## 核心理念
**业务部门不关心模型参数,只关心三件事:能省多少钱、能多赚多少钱、风险可不可控。**
所有输出必须围绕这个铁律组织。如果一句话里有技术名词但没有业务翻译,这句话就是噪音。
## 验收标准
每次输出完成后逐条自查:
View the complete source on GitHubc-drive-cleaner
When your agent uses it
Windows C盘清理和磁盘空间管理。当用户说C盘满了、磁盘空间不足、清理临时文件/缓存/回收站/系统日志、查找大文件、分析磁盘占用时使用。仅适用于 Windows 环境。不用于:macOS/Linux 磁盘清理、卸载软件(引导用户走系统卸载)、清理用户个人文件(只报告位置,删除决定权在用户)。
What it loads into context
Frontmatter
116 tokensAlways available to the agent
SKILL.md
1,126 tokensWhen the skill is activated
Reference documents
1,974 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 验收标准(清理任务完成前自查)
- 禁删清单(碰到即停)
- 工作流程
- 已知陷阱
- 权限与配套
- 脚本说明
Scripts executed separately
scripts/analyze_disk.pyscripts/clean_temp.pyscripts/find_large_files.py
Files
SKILL.mdcore3,821 bytesSKILL.mdcore3,821 bytesSKILL.mdcore3,821 bytesSKILL.mdcore3,821 bytesreferences/windows_cleanup_guide.mddoc6,151 bytesscripts/analyze_disk.pyruns3,937 bytesscripts/clean_temp.pyruns8,950 bytesscripts/find_large_files.pyruns5,678 bytes
Read SKILL.md
# C盘清理大师
安全导向的 Windows 磁盘清理:**分析 → 模拟 → 执行 → 验证**,破坏性操作永远先预览。
## 验收标准(清理任务完成前自查)
- [ ] 实际删除前完成过一次模拟运行,并把预览结果给用户看过
- [ ] `--execute` 前用户明确确认过(清理不经过回收站,不可恢复)
- [ ] 已提醒用户创建系统还原点和关闭浏览器(清缓存时)
- [ ] 未触碰"禁删清单"中的任何目录
- [ ] 清理后重跑分析,回报**实际释放的空间数字**,不是"应该清理了"
- [ ] 用户个人大文件(视频/压缩包等)只报告位置与大小,删除由用户自己决定
## 禁删清单(碰到即停)View the complete source on GitHubdeck-studio
When your agent uses it
PPT 生产 Agent:理解场景 → 推荐风格 → 先出大纲 → 定义页面 → 双通路交付(PPT 内容稿 / 逐页视觉图 / 信息图组图)。当用户要做汇报、路演、培训课件、提案、咨询报告的演示文稿,或要把大纲/文章做成 slides、deck、PPT、信息图、小红书图文、可视化传播图时使用。支持指定风格或从风格库推荐,支持 image 模型逐页出视觉稿。前身为 notion-infographic(v2)并吸收 ppt-designer 的设计原则。不用于:pptx 文件的纯技术操作(拆分/合并/提取,用宿主 pptx skill)、演讲稿或文章本体的写作(用 wechat-article-writer)、需要先做研究的内容调研(先走 deep-research 再回来)、单张海报或 logo 设计。
What it loads into context
Frontmatter
231 tokensAlways available to the agent
SKILL.md
3,631 tokensWhen the skill is activated
Reference documents
42,430 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 铁律(不是建议)
- 验收标准(交付前逐条自查)
- 不做什么
- 执行与真实性边界
- 工作流程(六步)
- 风格库(style-library/,当前 17 个;扩张按 `references/style-roadmap.md` 策展,终态 ≤14 正式 + 特例)
- 已知陷阱
- 参考文档(按需加载)
- 回归
Scripts executed separately
../../scripts/run_routing_evals.pyexamples/constructivist-design-constitution/generate.jsexamples/moshiro-consulting-report/generate.jsexamples/polar-night-ai-native/assemble.jsexamples/polar-night-ai-native/generate.jsexamples/yinghuang-bootcamp-proposal/assemble.jsexamples/yinghuang-bootcamp-proposal/generate.js
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore12,106 bytesevals/routing-evals.jsonasset2,413 bytesexamples/constructivist-design-constitution/README.mddoc2,033 bytesexamples/constructivist-design-constitution/generate.jsruns12,527 bytesexamples/constructivist-design-constitution/preview-cover.pngasset92,460 bytesexamples/moshiro-consulting-report/README.mddoc1,526 bytesexamples/moshiro-consulting-report/generate.jsruns15,351 bytesexamples/moshiro-consulting-report/preview-cover.pngasset88,454 bytesexamples/polar-night-ai-native/README.mddoc1,470 bytesexamples/polar-night-ai-native/assemble.jsruns1,558 bytesexamples/polar-night-ai-native/generate.jsruns16,599 bytesexamples/polar-night-ai-native/preview-cover.pngasset121,023 bytesexamples/yinghuang-bootcamp-proposal/README.mddoc1,584 bytesexamples/yinghuang-bootcamp-proposal/assemble.jsruns1,403 bytesexamples/yinghuang-bootcamp-proposal/generate.jsruns16,215 bytesexamples/yinghuang-bootcamp-proposal/preview-cover.pngasset71,813 bytesreferences/case-examples.mddoc13,677 bytesreferences/image-branch.mddoc2,633 bytesreferences/layout-registry.mddoc4,298 bytesreferences/layout-rules.mddoc14,338 bytesreferences/page-schema.mddoc2,903 bytesreferences/quick-reference.mddoc5,685 bytesreferences/scene-routing.mddoc2,364 bytesreferences/style-roadmap.mddoc3,460 bytesreferences/v2-pipeline/aesthetic-design.mddoc1,627 bytesreferences/v2-pipeline/design-system.mddoc6,910 bytesreferences/v2-pipeline/pptx-generation.mddoc6,287 bytesreferences/v2-pipeline/research-notes-format.mddoc3,554 bytesreferences/v2-pipeline/style-guide.mddoc4,864 bytesreferences/v2-pipeline/subagent-prompt.mddoc4,619 bytesreferences/v2-pipeline/v2-prompt-template.mddoc5,169 bytesreferences/v2-pipeline/v2-skill-archive.mddoc10,774 bytesreferences/visual-checklist.mddoc3,446 bytesstyle-library/business/aicher-preview.pngasset54,467 bytesstyle-library/business/aicher-system.mddoc1,801 bytesstyle-library/business/aicher-template.htmlasset2,583 bytesstyle-library/business/heibai-ledger.mddoc1,500 bytesstyle-library/business/moshiro-consulting.mddoc1,333 bytesstyle-library/creative/bauhaus-geometric.mddoc1,589 bytesstyle-library/creative/bauhaus-preview.pngasset56,359 bytesstyle-library/creative/bauhaus-template.htmlasset1,608 bytesstyle-library/creative/constructivist-preview.pngasset60,428 bytesstyle-library/creative/constructivist-red.mddoc1,627 bytesstyle-library/creative/constructivist-template.htmlasset1,581 bytesstyle-library/creative/tanghe-frame.mddoc1,403 bytesstyle-library/creative/yinghuang-studio.mddoc1,330 bytesstyle-library/custom/chaogeek-pixel.mddoc1,747 bytesstyle-library/custom/clawtime-industrial.mddoc1,509 bytesstyle-library/custom/workbuddy-modern.mddoc1,523 bytesstyle-library/education/academic-bluegray.mddoc1,424 bytesstyle-library/education/qingfeng-classroom.mddoc1,341 bytesstyle-library/media/hot-card.mddoc1,532 bytesstyle-library/media/neubrutalism-pop.mddoc1,550 bytesstyle-library/media/neubrutalism-preview.pngasset57,846 bytesstyle-library/media/neubrutalism-template.htmlasset1,762 bytesstyle-library/media/notion-handdrawn.mddoc1,630 bytesstyle-library/tech/platinum-future.mddoc1,402 bytesstyle-library/tech/polar-night.mddoc1,484 bytestemplates/image-prompt-template.mddoc1,213 bytestemplates/outline-template.mddoc852 bytestemplates/page-brief-template.mddoc867 bytes
Read SKILL.md
# Deck Studio:PPT 生产 Agent
不是"帮你排版润色",而是一条从场景理解到成品交付的生产线。
核心信条(承自 ppt-designer):**PPT 的本质是信息传达,设计是为内容服务的。**
## 铁律(不是建议)
1. **先 outline,后一切**。视觉图模式(模式 C)必须在大纲经用户确认后才启动;image 模型只负责"把已定义的页画出来",永远不决定 deck 讲什么。用户要求"跳过大纲直接画"时,解释原因并先给大纲。
2. **每页先有 page brief,再有 image prompt**,禁止跳级。
3. **一套 deck 一个风格**。风格参数从 style brief 统一注入每页,不允许逐页漂移。
## 验收标准(交付前逐条自查)
- [ ] outline 在任何页面产出之前存在且经用户确认(模式 A 除外,它本身就是终点)View the complete source on GitHubdeep-research
When your agent uses it
Use this skill when the user wants an evidence-based research memo, literature review, market/policy/technical landscape, or a multi-source decision brief with citations, trade-offs, and a clear conclusion. Best for tasks that need synthesis across multiple external sources, iterative follow-up research, or a reusable written artifact. Do not use for quick factual lookups, single-source summaries, simple Q&A, summarizing one document the user already provided, plan-only requests where the user explicitly defers the actual research, or when the user wants a short answer with no need for cited evidence (a short but evidence-backed conclusion is still a brief memo, which this skill covers). Chinese trigger examples: "帮我调研", "深度研究", "综述报告", "技术选型分析", "竞品研究", "政策分析".
What it loads into context
Frontmatter
297 tokensAlways available to the agent
SKILL.md
2,461 tokensWhen the skill is activated
Reference documents
7,940 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What this skill should produce
- When NOT to use this skill
- Org-policy boundary
- Active context bundle
- Workflow
- Deterministic helpers
- Evaluation and observability
- Degraded mode
- Stop conditions
Scripts executed separately
scripts/emit_run_summary.pyscripts/source_evaluator.pyscripts/verify_citations.py
Files
SKILL.mdcore11,264 bytesSKILL.mdcore11,264 bytesSKILL.mdcore11,264 bytesSKILL.mdcore11,264 bytesSKILL.mdcore11,264 bytesSKILL.mdcore11,264 bytesassets/report_template.mddoc1,006 bytesevals/routing-evals.jsonasset3,859 bytesevals/runbook.mddoc967 bytesreferences/evaluator-prompt.mddoc4,090 bytesreferences/handoff-format.mddoc4,027 bytesreferences/landscape-scan.mddoc1,194 bytesreferences/methodology.mddoc5,852 bytesreferences/observability.mddoc3,051 bytesreferences/quality-gates.mddoc5,593 bytesreferences/report-assembly.mddoc3,435 bytesreferences/research-notes-format.mddoc2,203 bytesreferences/subagent-prompt.mddoc1,787 bytesreferences/tension-discovery.mddoc1,235 bytesscripts/emit_run_summary.pyruns5,129 bytesscripts/source_evaluator.pyruns12,234 bytesscripts/verify_citations.pyruns11,006 bytes
Read SKILL.md
# Deep Research V8.1
This skill is for **evidence-rich research outputs**, not for every question that happens to mention “analysis”.
The V8 shift is simple:
- **Single-agent first.** Start with one lead agent and only fan out when parallel work will clearly help.
- **Thin harness, fat skill.** Put reusable judgment and workflow here; keep deterministic checks in scripts.
- **Context organization over prompt stuffing.** Load the minimum active context bundle, then pull in references only when needed.
- **Eval and observability built in.** A good report is not enough; the run must also be diagnosable and improvable.
## What this skill should produce
Choose the lightest artifact that satisfies the task.
View the complete source on GitHubkeqian-method
When your agent uses it
胥克谦式AI-Native产品开发方法论。适用于:(1) 使用AI Agent(Claude Code、Codex、Cursor等)进行产品级软件开发,(2) 设计和优化Harness/Skill体系,(3) 文档驱动开发(SDD)流程,(4) 构建自动化质量门禁和eval机制,(5) Token成本优化与缓存策略,(6) 产品人转型开发者的AI编程实践。触发场景包括"帮我设计开发流程"、"怎么降低token成本"、"怎么提高AI编码质量"、"文档驱动"、"质量门禁"、"harness设计"、"单agent vs multi-agent"、"自动化迭代"、"AI产品开发"、"SDD"、"eval机制"等。即使用户只是说"帮我用AI写代码"或"怎么让agent干活更靠谱"也应触发。注意:如果产品是行为开放、用户输入不可穷举的AI-native类型,请改用 xuefeng-method skill。不用于:单个bug修复或小改动(无需方法论)、PRD需求文档写作(用product-manager)。
What it loads into context
Frontmatter
284 tokensAlways available to the agent
SKILL.md
2,691 tokensWhen the skill is activated
Reference documents
2,039 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 第一原则:Iron Law(铁律)
- 第二原则:单Agent极致论
- 第三原则:文档驱动开发(SDD)
- 第四原则:质量门禁闭环(Verification-Driven)
- 第五原则:产品拆解思维
- 第六原则:与AI斗智斗勇
- 实战工作流模板
- 模型选择建议
- 成本优化清单
- 心法总结
- 验收标准(按本方法论执行的任务,交付前自查)
- 参考资料
Scripts executed separately
../../scripts/run_routing_evals.py
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore9,004 bytesevals/routing-evals.jsonasset1,816 bytesreferences/eval-patterns.mddoc4,058 bytesreferences/sdd-framework.mddoc2,776 bytes
Read SKILL.md
# 克谦方法论:AI-Native产品开发实战体系
> 核心理念:产品人思维 × 极致单Agent × 文档驱动 × 质量门禁闭环
>
> 来源:胥克谦——从音乐教师到产品经理到AI-Native连续创业者,皮影客创始人,
> 十几万行自建skill和脚本的harness工程实践者。
---
## 第一原则:Iron Law(铁律)
**概率乘是第一性原理。**
每个环节的成功率相乘决定最终质量。即使每次0.99,n=51后也不及格。View the complete source on GitHubmineru-pdf-parser
When your agent uses it
用 MinerU 将复杂PDF文档转换为LLM友好的Markdown/JSON格式。适用于:(1) PDF转Markdown/JSON,(2) 提取PDF中的文本、表格、公式、图像,(3) 解析学术论文、技术文档、商业报告,(4) 为RAG应用准备文档数据,(5) 批量处理PDF。触发关键词:"PDF解析"、"PDF转Markdown"、"提取PDF表格/公式"、"MinerU"、"parse PDF"等。不用于:PDF的阅读/填表/签名/拆分合并(用宿主pdf工具)、Word/PPT等非PDF格式解析、只需读几页内容的场景(直接读即可,不必转换)。
What it loads into context
Frontmatter
187 tokensAlways available to the agent
SKILL.md
1,137 tokensWhen the skill is activated
Reference documents
3,587 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 安装
- 快速使用
- 解析模式选择
- 输出文件
- 最佳实践
- 验收标准(解析任务完成前自查)
- 已知陷阱
- 脚本
Scripts executed separately
scripts/mineru_parse.py
Files
SKILL.mdcore3,919 bytesSKILL.mdcore3,919 bytesSKILL.mdcore3,919 bytesSKILL.mdcore3,919 bytesSKILL.mdcore3,919 bytesSKILL.mdcore3,919 bytesSKILL.mdcore3,919 bytesreferences/api_reference.mddoc3,076 bytesreferences/best_practices.mddoc5,680 bytesreferences/output_formats.mddoc3,202 bytesscripts/mineru_parse.pyruns8,367 bytes
Read SKILL.md
# MinerU PDF Parser
将复杂PDF文档转换为机器可读的Markdown/JSON格式,适用于LLM和RAG应用。
## 安装
```bash
# 推荐使用uv安装
pip install uv
uv pip install -U "mineru[all]"
# 下载模型(首次使用)
mineru-models-download
```View the complete source on GitHubpair-programming
When your agent uses it
结对编程搭档。当用户要求"边写边审"、"结对编程"、"写完自己 review 一遍"、"高可靠地实现",或明确希望代码交付时附带自我审查意见时使用。交付代码的同时输出结构化审查(正确性/安全/性能/可读性/健壮性五维度),重点捕捉 AI 生成代码的特有缺陷。不用于:对已有 PR 的正式评审(用 code review 流程)、安全专项扫描(用 security-audit)、10 行以内的简单片段。
What it loads into context
Frontmatter
150 tokensAlways available to the agent
SKILL.md
1,171 tokensWhen the skill is activated
Reference documents
1,259 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 验收标准(每次交付前自查)
- 不做什么
- 审查工作流
- AI 生成代码的特有缺陷(重点盯防)
- 反馈格式与语气
- 引用资源
Scripts executed separately
../../scripts/run_routing_evals.py
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore4,028 bytesevals/routing-evals.jsonasset1,566 bytesreferences/detailed-checklist.mddoc4,392 bytes
Read SKILL.md
# 结对编程搭档
交付代码 + 像负责任的高级开发者一样自我审查,一次给到位。
## 验收标准(每次交付前自查)
- [ ] 代码之后附有结构化审查意见(好的地方 / 需要关注 / 优化建议三段)
- [ ] 每个"需要关注"都给了可执行的修改方案,不是只指出问题
- [ ] 🔴 必查项五条全部过检(空值/输入验证/注入/敏感数据/资源泄漏)
- [ ] 审查意见针对**本次生成的代码**,不是通用清单复读
- [ ] 发现自己代码的问题时直接修掉再交付,审查意见只留真正的权衡点
## 不做什么
View the complete source on GitHubproduct-manager
When your agent uses it
资深产品经理助手,提供 PRD/MRD/BRD 创作与评审、产品策略、留存增长、竞品分析、功能优先级,以及 grill-me-to-doc 逐轮访谈。用户要求“逐个问我”“先把需求问清楚”“grill me”“把想法变成产品文档”时进入 grill-me-to-doc:先读仓库证据,每轮只问一个决策,给推荐答案与理由,记录决策和未决项,支持 resume,产出结构化 PRODUCT-DOC;文档完成且用户批准前硬停止,任何时候都不写实现代码。即使未提“产品”,讨论 App 功能、增长或商业模式也应触发。不用于:单纯代码实现、系统架构深设(用 solution-architect)、需多源引用的市场调研(用 deep-research)、纯营销文案。
What it loads into context
Frontmatter
213 tokensAlways available to the agent
SKILL.md
5,255 tokensWhen the skill is activated
Reference documents
18,558 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 上下文感知原则
- 工作模式
- 不做什么
- 已知陷阱
- 通用能力
- 产品思维方法论
- 资源文档
Scripts executed separately
../../scripts/run_routing_evals.pyscripts/validate_grill_session.py
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore18,824 bytesevals/routing-evals.jsonasset2,180 bytesreferences/GRILL-ME-TO-DOC.mddoc6,117 bytesreferences/PM-BEST-PRACTICES.mddoc17,094 bytesreferences/PRD-TEMPLATE.mddoc11,880 bytesreferences/PRD-WRITING-GUIDE.mddoc13,288 bytesreferences/PRODUCT-DOC-TEMPLATE.mddoc1,645 bytesreferences/REVIEW-CHECKLIST.mddoc13,094 bytesschemas/grill-state.schema.jsonasset5,148 bytesscripts/validate_grill_session.pyruns12,564 bytes
Read SKILL.md
# Product Manager Skill
## 概述
资深产品经理能力,覆盖三大核心场景:文档创作与评审、产品策略咨询、竞品与市场研究。skill的价值在于提供系统化的分析框架和场景化的深度建议,而非机械套用模板。
## 上下文感知原则
在开始工作前,先评估用户已经提供了多少信息,据此决定行为模式:
**信息充足**(用户已给出产品名称、目标用户、核心功能、技术栈等关键信息)→ 直接开始工作,在过程中补充追问。不要一上来就问一堆问题让用户等待。
**信息部分缺失**(有基本方向但缺关键细节)→ 先开始工作产出初稿框架,在关键决策点标注待确认项,最后集中提问1-2个最关键的问题。
View the complete source on GitHubsecurity-audit
When your agent uses it
全面的代码安全检查和服务器安全审计skill。适用于:(1) 代码漏洞扫描 - 检测SQL注入、XSS、SSRF等OWASP Top 10漏洞,(2) 依赖安全检查 - 识别过时或有漏洞的第三方库,结合实时搜索确认最新CVE,(3) 服务器配置审计 - 检查SSH、防火墙、权限等安全配置,(4) 敏感信息泄露检测 - API密钥、密码、令牌等硬编码检测,(5) 容器安全扫描 - Docker镜像和Kubernetes配置审计,(6) CI/CD安全检查。触发关键词:"安全检查"、"漏洞扫描"、"代码审计"、"security audit"、"vulnerability scan"、"SAST"、"dependency check"、"CVE检测"等。不用于:修复单个已定位的bug、编写新的安全功能代码、对无授权的第三方系统做扫描或渗透测试。
What it loads into context
Frontmatter
242 tokensAlways available to the agent
SKILL.md
2,539 tokensWhen the skill is activated
Reference documents
6,995 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 授权与边界(先确认再动手)
- 时效性规则(重要)
- 工作流程
- 快速开始
- 检测能力
- 扫描工具使用
- 服务器安全检查
- 修复建议
- 报告格式
- 验收标准(完成前逐条自查)
- 已知陷阱
- 最佳实践
- 参考资料
Scripts executed separately
scripts/dependency_check.pyscripts/full_scan.pyscripts/secrets_scan.py
Files
SKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesSKILL.mdcore8,834 bytesevals/routing-evals.jsonasset1,482 bytesreferences/detection_rules.mddoc6,575 bytesreferences/remediation_guide.mddoc8,781 bytesreferences/server_hardening.mddoc7,222 bytesscripts/dependency_check.pyruns12,584 bytesscripts/full_scan.pyruns19,486 bytesscripts/secrets_scan.pyruns15,810 bytes
Read SKILL.md
# Security Audit Skill
全面的安全审计工具,覆盖代码静态分析(SAST)、依赖检查(SCA)和服务器配置审计。
## 授权与边界(先确认再动手)
- 只审计用户拥有或明确获得授权的代码/服务器;对第三方系统的扫描请求一律拒绝并说明原因
- 本skill做**审计与建议**,不做主动攻击性测试(漏洞利用、爆破、DoS验证)
- 发现硬编码凭证时只报告**位置与类型**,不在输出中回显完整明文
- 除非用户明确要求,不直接修改业务代码;修复以建议形式给出
## 时效性规则(重要)
本文件**不维护具体CVE清单**。凡涉及"某版本是否有漏洞"的结论,必须现场查询:View the complete source on GitHubsolution-architect
When your agent uses it
解决方案架构师助手。当用户要设计新系统架构、评审现有架构、做技术选型决策、诊断性能/可扩展性/可用性问题、规划架构演进或重构时使用。覆盖微服务、事件驱动、云原生等架构模式,技术趋势通过实时搜索获取而非依赖内置知识。不用于:具体功能的代码实现、安全漏洞审计(用security-audit)、产品需求分析(用product-manager)、只需一句话回答的技术常识问题。
What it loads into context
Frontmatter
138 tokensAlways available to the agent
SKILL.md
1,630 tokensWhen the skill is activated
Reference documents
14,373 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 验收标准(交付前逐条自查)
- 不做什么
- 工作流程
- 时效性规则
- 核心原则:架构设计的"黄金法则"
- 已知陷阱
- 参考文档(按需加载)
Scripts executed separately
../../scripts/run_routing_evals.py
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore5,689 bytesevals/routing-evals.jsonasset1,653 bytesreferences/场景与输出模板.mddoc3,607 bytesreferences/技术选型指南.mddoc9,132 bytesreferences/架构模式.mddoc13,646 bytesreferences/架构设计原则.mddoc11,003 bytesreferences/架构评审清单.mddoc7,969 bytes
Read SKILL.md
# 解决方案架构师
帮助设计高质量的软件系统架构,做出明智的技术选型决策,提供架构评审和优化建议。
## 验收标准(交付前逐条自查)
- [ ] 需求先行:输出前已确认业务需求、非功能需求(性能/可用性/安全/扩展性指标)和约束条件(预算/团队/技术栈)
- [ ] 每个关键技术选择都有明确理由和被否掉的替代方案
- [ ] 权衡透明:说清方案的代价和风险,没有"银弹"式推荐
- [ ] 涉及"最新/流行/趋势"的论断有本次搜索的依据,不是训练数据里的旧闻
- [ ] 建议与团队能力匹配(不给 5 人小团队推 30 个微服务)
- [ ] 重要决策以 ADR 形式可记录,给出下一步行动(POC/预研/评审项)
## 不做什么View the complete source on GitHubthreejs-performance
When your agent uses it
Three.js 性能优化指南。当 Three.js/React Three Fiber 项目出现掉帧、内存增长、加载缓慢、绘制调用过多,或需要迁移 WebGPU、实现大规模粒子/实例化渲染、配置资源压缩管线时使用。触发关键词:"Three.js 优化"、"WebGPU"、"draw calls"、"内存泄漏"、"TSL"、"InstancedMesh"、"R3F 优化"等。不用于:通用 React 性能问题(用 vercel-react-best-practices)、WebGL/Three.js 入门教学、3D 美术资产的制作本身。
What it loads into context
Frontmatter
163 tokensAlways available to the agent
SKILL.md
1,814 tokensWhen the skill is activated
Reference documents
4,853 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 黄金法则
- 诊断决策树
- 各领域规则速查
- 验收标准(优化任务完成前自查)
- 不做什么
- 已知陷阱
- 调试工具
- 参考文档(按需加载)
Files
SKILL.mdcore5,957 bytesreferences/assets.mddoc4,245 bytesreferences/examples.mddoc10,190 bytesreferences/webgpu.mddoc3,637 bytes
Read SKILL.md
# Three.js 性能优化指南
## 黄金法则
**绘制调用 < 100 次/帧。** 三角形数量不如绘制调用数量重要;超过 500 次绘制调用,强大 GPU 也会吃力。用 `renderer.info.render.calls` 监控(完整监控代码见 `references/examples.md`)。
## 诊断决策树
| 症状 | 先查 | 深入 |
|------|------|------|
| 掉帧、卡顿 | `renderer.info.render.calls` 是否 >100 | 实例化/合批(下文),后处理链 |
| 内存持续增长 | `renderer.info.memory` 的 geometries/textures 计数是否只增不减 | dispose 铁律(下文),`references/examples.md` 完整清理代码 |
| 加载慢、显存爆 | 模型是否未压缩 | `references/assets.md` 压缩管线 |
| 粒子/物理瓶颈 | CPU 粒子是否 >5 万 | WebGPU 计算着色器,`references/webgpu.md` |View the complete source on GitHubwechat-article-writer
When your agent uses it
专业微信公众号文章助手,支持四个独立且可组合模式:article 写正文;image-prompts 从文章生成版本化、provider-neutral 的图片提示词 manifest 与稳定占位符,但不调用生图;layout 把文章和 manifest 确定性转换为微信安全的内联 HTML;full-pipeline 串联正文、提示词/占位符和排版。用户提到公众号、微信文章、推文、配图提示词、图片 manifest、公众号排版、内联 HTML 或完整图文流水线时触发。支持企业官号、个人技术博客、活动回顾和产品评测。不用于:带引用的研究报告(用 deep-research)、PPT/组图(用 deck-studio)、英文文章、仅润色一小段。永不自动发布,永不把提示词或占位符声称为已生成图片。
What it loads into context
Frontmatter
213 tokensAlways available to the agent
SKILL.md
4,043 tokensWhen the skill is activated
Reference documents
16,177 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 先选执行模式
- 第一步:风格识别与素材分析
- 第二步:内容研究
- 第三步:按风格创作
- 第四步:标题创作
- 第五步:排版与交付
- 安全与真实性红线
- 第六步:质量自检
- 第七步(可选):L4 反翻译腔 Polish
- 资源文档
Scripts executed separately
../../scripts/run_routing_evals.pyscripts/render_wechat_layout.pyscripts/validate_image_manifest.py
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore14,342 bytesSKILL.mdcore14,342 bytesSKILL.mdcore14,342 bytesSKILL.mdcore14,342 bytesassets/emoji-library.txtdoc8,539 bytesassets/template-structure.mddoc8,514 bytesevals/routing-evals.jsonasset2,883 bytesreferences/PIPELINE-CONTRACT.mddoc4,386 bytesreferences/linguistic-background.mddoc3,247 bytesreferences/quality-checklist.mddoc10,167 bytesreferences/style-guide.mddoc10,381 bytesreferences/title-formulas.mddoc8,528 bytesreferences/writing-techniques.mddoc7,395 bytesschemas/image-manifest.schema.jsonasset3,166 bytesscripts/render_wechat_layout.pyruns6,610 bytesscripts/validate_image_manifest.pyruns8,327 bytes
Read SKILL.md
# 微信公众号文章创作助手
## 概述
帮助用户创作高质量的微信公众号文章。核心能力是**风格灵活适配**——同一个skill既能写企业官号的正式文案,也能写个人IP的技术博客。根据用户需求自动匹配最合适的风格和结构。
## 先选执行模式
用户未指定时默认 `article`。需要跨模式时按
`references/PIPELINE-CONTRACT.md` 的文件接口交接,不靠对话隐含状态。
| 模式 | 输入 | 必须输出 | 停止位置 |
|---|---|---|---|
| `article` | 素材/选题 | `article.md` | 正文完成 |View the complete source on GitHubxuefeng-method
When your agent uses it
雪峰式AI-Native产品开发方法论。适用于:(1) 用户行为开放、不可穷举的AI-native产品(AI日历、AI助手、AI推荐、对话式产品等),(2) 强模型依赖型场景,AI驱动核心决策而非仅辅助,(3) 多专精Agent架构设计与分工,(4) 上线后快速校准、行为审计与漂移检测,(5) 模型选择和智能路由策略,(6) 概率性输出的质量评估。触发场景包括"AI-native产品怎么做"、"用户行为不可预测怎么办"、"多agent怎么分工"、"模型漂移怎么处理"、"校准到95%太难了"、"唯快不破"、"怎么选模型"、"agent并行分工"、"AI产品上线后怎么迭代"。注意:如果产品是场景明确、边界可定义的+AI类型,请改用 keqian-method skill。即使用户没有明确说"AI-native",但在讨论AI驱动决策、用户行为不可预测、概率性输出等话题时也应触发。
What it loads into context
Frontmatter
285 tokensAlways available to the agent
SKILL.md
3,191 tokensWhen the skill is activated
Reference documents
3,412 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 第零步:产品类型判断(必须先做)
- 第一原则:穷举是死循环
- 第二原则:多养专精虾
- 第三原则:唯快不破
- 第四原则:模型选择实战
- 第五原则:行为审计替代质量门禁
- 第六原则:一切以用户为中心
- 验收标准(按本方法论执行的任务,交付前自查)
- 实战与协作模板
- 参考资料
Scripts executed separately
../../scripts/run_routing_evals.py
Files
../../scripts/run_routing_evals.pyruns6,893 bytesSKILL.mdcore11,204 bytesSKILL.mdcore11,204 bytesevals/routing-evals.jsonasset1,822 bytesreferences/behavioral-clusters.mddoc3,491 bytesreferences/drift-detection.mddoc4,532 bytesreferences/model-routing.mddoc1,129 bytesreferences/operating-playbook.mddoc2,170 bytes
Read SKILL.md
# 雪峰方法论:AI-Native 产品开发实战体系
> 核心理念:强模型依赖 × 多专精Agent × 快速校准 × 行为审计
>
> 来源:雪峰——AI-Native连续创业者,深耕AI日历管理等AI驱动产品。
> 核心洞察:穷举是死循环,唯快不破才是AI-native的生存之道。
>
> 与克谦方法论(keqian-method)互为对偶:
> 克谦解决"如何让AI在明确边界内可靠执行",
> 雪峰解决"当边界本身不确定时怎么办"。
---
## 第零步:产品类型判断(必须先做)View the complete source on GitHubSkills included
- a-share-analyst
- gaokao-expert
- podcast-generator
- seedream-imagegen
- university-exam-prep
- weather-forecast-report
- llm-wiki
- ai-sales-champion
- c-drive-cleaner
- deck-studio
- deep-research
- keqian-method
- mineru-pdf-parser
- pair-programming
- product-manager
- security-audit
- solution-architect
- threejs-performance
- wechat-article-writer
- xuefeng-method
Repository facts
- License
- MIT
- Created
- Oct 31, 2025
- Last pushed
- Aug 13, 2026
- Forks
- 129
- Open issues
- 1
- Project activity
- Slowing
Rank over the last 12 weeks
Now #901 · best #695 · down 71 in 30 days
- #852 on Oct 9
- #871 on Oct 15
- #887 on Oct 19
- #895 on Oct 20
- #913 on Oct 21
- #901 since Oct 23
31 earlier periods
- #696 on Aug 2
- #741 on Aug 6
- #708 on Aug 9
- #732 on Aug 10
- #736 on Aug 12
- #738 on Aug 13
- #702 on Aug 15
- #719 on Aug 16
- #726 on Aug 17
- #743 on Aug 18
- #731 on Aug 21
- #695 on Aug 22
- #753 on Aug 24
- #787 on Aug 28
- #764 on Sep 2
- #776 on Sep 3
- #735 on Sep 5
- #764 on Sep 6
- #777 on Sep 7
- #797 on Sep 8
- #774 on Sep 12
- #790 on Sep 15
- #805 on Sep 16
- #809 on Sep 17
- #830 on Sep 22
- #758 on Sep 25
- #795 on Sep 27
- #851 on Sep 29
- #779 on Oct 1
- #773 on Oct 3
- #805 on Oct 4
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 5, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 33 notes
lab/Geek-skills-a-share-analyst/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedlab/Geek-skills-a-share-analyst/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedlab/Geek-skills-podcast-generator/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedlab/Geek-skills-podcast-generator/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedlab/Geek-skills-seedream-imagegen/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedllm-wiki/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-c-drive-cleaner/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-c-drive-cleaner/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-c-drive-cleaner/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-deep-research/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-deep-research/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-deep-research/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-deep-research/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-deep-research/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-mineru-pdf-parser/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-mineru-pdf-parser/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-mineru-pdf-parser/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-mineru-pdf-parser/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-mineru-pdf-parser/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/Geek-skills-mineru-pdf-parser/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- 58 dangerous shell commandsin 3 files: Geek-skills-security-audit/references/server_hardening.md:51 Geek-skills-security-audit/references/detection_rules.md:85 Geek-skills-security-audit/SKILL.md:136
- 49 network referencesin 16 files: Geek-skills-seedream-imagegen/scripts/generate_image.py:34 Geek-skills-security-audit/scripts/secrets_scan.py:62 Geek-skills-mineru-pdf-parser/references/api_reference.md:148 and 13 more
- 1 obfuscation signalin 1 file: Geek-skills-wechat-article-writer/assets/emoji-library.txt:60
- No instruction overrides
- 41 executable scriptsin 30 files: scripts/run_routing_evals.py:1 Geek-skills-a-share-analyst/scripts/fetch_market_data.py:1 Geek-skills-a-share-analyst/scripts/generate_report.py:1 and 27 more
- 41 references to secretsin 10 files: Geek-skills-seedream-imagegen/scripts/generate_image.py:21 Geek-skills-seedream-imagegen/references/quick_start.md:20 Geek-skills-c-drive-cleaner/scripts/clean_temp.py:69 and 7 more
- No requests to vote in Hot
Show all 190 signals by file
lab/Geek-skills-a-share-analyst/scripts/fetch_market_data.py
- Executable scriptline 1
#!/usr/bin/env python3
lab/Geek-skills-a-share-analyst/scripts/generate_report.py
- Executable scriptline 1
#!/usr/bin/env python3
lab/Geek-skills-a-share-analyst/scripts/strategy_multi_factor.py
- Executable scriptline 1
#!/usr/bin/env python3
lab/Geek-skills-a-share-analyst/scripts/technical_analysis.py
- Executable scriptline 1
#!/usr/bin/env python3
lab/Geek-skills-gaokao-expert/scripts/analyze_paper.py
- Executable scriptline 1
#!/usr/bin/env python3
lab/Geek-skills-gaokao-expert/scripts/analyze_question.py
- Executable scriptline 1
#!/usr/bin/env python3
lab/Geek-skills-podcast-generator/SKILL.md
lab/Geek-skills-podcast-generator/references/api_reference.md
- Network accessline 17
**Get credentials:** https://console.volcengine.com/speech/service/10028
lab/Geek-skills-podcast-generator/scripts/generate_podcast.py
lab/Geek-skills-seedream-imagegen/SKILL.md
lab/Geek-skills-seedream-imagegen/references/api_reference.md
- Network accessline 7
**API 端点**: `https://api.segmind.com/v1/seedream-4`
lab/Geek-skills-seedream-imagegen/references/quick_start.md
- Network accessline 11
### 获取 API 密钥 1. 访问 https://www.segmind.com/ - Network accessline 12
1. 访问 https://www.segmind.com/ - Access to secretsline 20
[Credential-related source omitted] - Access to secretsline 41
[Credential-related source omitted] - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 48
[Credential-related source omitted] - Access to secretsline 52
[Credential-related source omitted] - Access to secretsline 57
[Credential-related source omitted] - Access to secretsline 60
[Credential-related source omitted] - Access to secretsline 69
[Credential-related source omitted]
lab/Geek-skills-seedream-imagegen/scripts/generate_image.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 21
[Credential-related source omitted] - Access to secretsline 26
[Credential-related source omitted] - Access to secretsline 28
[Credential-related source omitted] - Access to secretsline 29
[Credential-related source omitted] - Access to secretsline 31
[Credential-related source omitted] - Network accessline 34
self.api_url = "https://api.segmind.com/v1/seedream-4" - Access to secretsline 36
[Credential-related source omitted] - Network accessline 99
try: response = requests.post( - Network accessline 100
response = requests.post( - Network accessline 175
filename = f"seedream_{timestamp}_{prompt_short}_{i+1}.png" filepath = output_path / filename img_response = requests.get(img_url, timeout=30) - Network accessline 176
filepath = output_path / filename img_response = requests.get(img_url, timeout=30) with open(filepath, 'wb') as f: - Network accessline 177
img_response = requests.get(img_url, timeout=30) - Access to secretsline 202
[Credential-related source omitted] - Access to secretsline 203
[Credential-related source omitted] - Access to secretsline 204
[Credential-related source omitted] - Access to secretsline 205
[Credential-related source omitted] - Access to secretsline 209
[Credential-related source omitted] - Access to secretsline 307
[Credential-related source omitted] - Access to secretsline 362
[Credential-related source omitted]
lab/Geek-skills-university-exam-prep/scripts/analyze_material.py
- Executable scriptline 1
#!/usr/bin/env python3
lab/Geek-skills-university-exam-prep/scripts/generate_questions.py
- Executable scriptline 1
#!/usr/bin/env python3
llm-wiki/SKILL.md
- Network accessline 10
Based on [Andrej Karpathy's LLM Wiki pattern](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f): raw sources are "compiled" into a structured wiki that compounds over time.
scripts/run_routing_evals.py
- Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3 - Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-c-drive-cleaner/references/windows_cleanup_guide.md
- Network accessline 209
- [Microsoft官方磁盘清理指南](https://support.microsoft.com/windows) - Network accessline 210
- [DISM命令参考](https://docs.microsoft.com/windows-hardware/manufacture/desktop/dism) - Network accessline 211
- [磁盘清理最佳实践](https://docs.microsoft.com/troubleshoot/windows-server/backup-and-storage/disk-cleanup-tool)
skills/Geek-skills-c-drive-cleaner/scripts/analyze_disk.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-c-drive-cleaner/scripts/clean_temp.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 69
[Credential-related source omitted] - Access to secretsline 70
[Credential-related source omitted] - Access to secretsline 72
[Credential-related source omitted] - Access to secretsline 149
[Credential-related source omitted] - Access to secretsline 154
[Credential-related source omitted]
skills/Geek-skills-c-drive-cleaner/scripts/find_large_files.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-deck-studio/examples/constructivist-design-constitution/generate.js
- Executable scriptline 1
Executable file
skills/Geek-skills-deck-studio/examples/moshiro-consulting-report/generate.js
- Executable scriptline 1
Executable file
skills/Geek-skills-deck-studio/examples/polar-night-ai-native/assemble.js
- Executable scriptline 1
Executable file
skills/Geek-skills-deck-studio/examples/polar-night-ai-native/generate.js
- Executable scriptline 1
Executable file
skills/Geek-skills-deck-studio/examples/yinghuang-bootcamp-proposal/assemble.js
- Executable scriptline 1
Executable file
skills/Geek-skills-deck-studio/examples/yinghuang-bootcamp-proposal/generate.js
- Executable scriptline 1
Executable file
skills/Geek-skills-deck-studio/references/v2-pipeline/research-notes-format.md
- Network accessline 30
[1] GitHub Copilot 用户数据报告 | https://... | 权威性: 9/10 | 2025-03 - Network accessline 31
[2] Stack Overflow 2025 开发者调查 | https://... | 权威性: 8/10 | 2025-06 - Network accessline 32
[3] Gartner AI 编程工具魔力象限 | https://... | 权威性: 9/10 | 2025-04 - Network accessline 33
[4] 某创业公司CTO的博客 | https://... | 权威性: 5/10 | 2025-05
skills/Geek-skills-deep-research/scripts/emit_run_summary.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-deep-research/scripts/source_evaluator.py
skills/Geek-skills-deep-research/scripts/verify_citations.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-mineru-pdf-parser/references/api_reference.md
- Network accessline 148
启动服务后访问 `http://127.0.0.1:8000/docs` 查看完整Swagger文档。 - Network accessline 156
访问 `http://127.0.0.1:7860` 使用图形界面。 - Network accessline 167
[Credential-related source omitted] - Network accessline 168
[Credential-related source omitted] - Access to secretsline 168
[Credential-related source omitted] - Network accessline 169
"base_url": "https://api.openai.com/v1",
skills/Geek-skills-mineru-pdf-parser/references/best_practices.md
- Network accessline 135
python -m pip install paddlepaddle-gpu==3.0.0b1 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/
skills/Geek-skills-mineru-pdf-parser/scripts/mineru_parse.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-product-manager/schemas/grill-state.schema.json
skills/Geek-skills-product-manager/scripts/validate_grill_session.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-security-audit/SKILL.md
skills/Geek-skills-security-audit/references/detection_rules.md
- Dangerous shell operationline 85
exec("ls " + userInput); - Network accessline 129
# Python requests.get(user_url) - Network accessline 130
requests.get(user_url) - Network accessline 131
urllib.request.urlopen(user_provided_url) - Network accessline 141
// Node.js fetch(userUrl); - Network accessline 142
fetch(userUrl); - Dangerous shell operationline 179
eval(user_input) - Dangerous shell operationline 188
eval(userInput);
skills/Geek-skills-security-audit/references/remediation_guide.md
- Access to secretsline 36
[Credential-related source omitted] - Access to secretsline 113
[Credential-related source omitted] - Network accessline 379
# 使用示例 if is_safe_url(user_url): response = requests.get(user_url) - Network accessline 380
if is_safe_url(user_url): response = requests.get(user_url) else: - Network accessline 381
response = requests.get(user_url)
skills/Geek-skills-security-audit/references/server_hardening.md
- Dangerous shell operationline 51
sudo cp /etc/ssh/sshd_config /etc/ssh/sshd_config.backup - Dangerous shell operationline 54
sudo sshd -t - Dangerous shell operationline 57
sudo systemctl restart sshd - Dangerous shell operationline 68
sudo ufw enable - Dangerous shell operationline 71
sudo ufw default deny incoming - Dangerous shell operationline 72
sudo ufw default allow outgoing - Dangerous shell operationline 75
sudo ufw allow 2222/tcp - Dangerous shell operationline 78
sudo ufw allow 80/tcp - Dangerous shell operationline 79
sudo ufw allow 443/tcp - Dangerous shell operationline 82
sudo ufw limit 2222/tcp - Dangerous shell operationline 85
sudo ufw status verbose - Dangerous shell operationline 92
sudo iptables -F - Dangerous shell operationline 93
sudo iptables -X - Dangerous shell operationline 96
sudo iptables -P INPUT DROP - Dangerous shell operationline 97
sudo iptables -P FORWARD DROP - Dangerous shell operationline 98
sudo iptables -P OUTPUT ACCEPT - Dangerous shell operationline 101
sudo iptables -A INPUT -i lo -j ACCEPT - Dangerous shell operationline 104
sudo iptables -A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT - Dangerous shell operationline 107
sudo iptables -A INPUT -p tcp --dport 2222 -j ACCEPT - Dangerous shell operationline 110
sudo iptables -A INPUT -p tcp --dport 80 -j ACCEPT - Dangerous shell operationline 111
sudo iptables -A INPUT -p tcp --dport 443 -j ACCEPT - Dangerous shell operationline 114
sudo iptables -A INPUT -p tcp --syn -m limit --limit 1/s -j ACCEPT - Dangerous shell operationline 117
sudo iptables-save > /etc/iptables/rules.v4 - Dangerous shell operationline 128
sudo adduser deploy - Dangerous shell operationline 131
sudo usermod -aG sudo deploy - Dangerous shell operationline 134
echo "deploy ALL=(ALL) NOPASSWD:ALL" | sudo tee /etc/sudoers.d/deploy - Access to secretsline 144
[Credential-related source omitted] - Access to secretsline 147
[Credential-related source omitted] - Access to secretsline 148
[Credential-related source omitted] - Dangerous shell operationline 159
sudo find / -perm -4000 -type f 2>/dev/null - Dangerous shell operationline 162
sudo find / -perm -2000 -type f 2>/dev/null - Dangerous shell operationline 165
sudo find / -perm -002 -type f 2>/dev/null - Dangerous shell operationline 168
sudo find / -nouser -o -nogroup 2>/dev/null - Dangerous shell operationline 171
sudo find / -name ".rhosts" 2>/dev/null - Dangerous shell operationline 178
sudo chmod 755 /etc - Dangerous shell operationline 179
sudo chmod 644 /etc/passwd - Dangerous shell operationline 180
sudo chmod 640 /etc/shadow - Dangerous shell operationline 181
sudo chmod 644 /etc/group - Dangerous shell operationline 184
sudo chmod 700 /home/* - Dangerous shell operationline 187
sudo chmod 1777 /tmp - Dangerous shell operationline 188
sudo chmod 1777 /var/tmp - Dangerous shell operationline 207
sudo apt update - Dangerous shell operationline 208
sudo apt install unattended-upgrades - Dangerous shell operationline 209
sudo dpkg-reconfigure unattended-upgrades - Dangerous shell operationline 221
sudo apt update && sudo apt upgrade -y - Dangerous shell operationline 224
sudo needrestart -r a - Dangerous shell operationline 235
sudo apt install auditd audispd-plugins - Dangerous shell operationline 238
sudo systemctl enable auditd - Dangerous shell operationline 239
sudo systemctl start auditd - Dangerous shell operationline 250
sudo auditctl -R /etc/audit/rules.d/audit.rules - Dangerous shell operationline 281
sudo apt install fail2ban - Dangerous shell operationline 284
sudo cp /etc/fail2ban/jail.conf /etc/fail2ban/jail.local - Dangerous shell operationline 327
sudo systemctl disable cups - Dangerous shell operationline 328
sudo systemctl disable avahi-daemon - Dangerous shell operationline 329
sudo systemctl disable bluetooth - Dangerous shell operationline 363
sudo sysctl -p
skills/Geek-skills-security-audit/scripts/dependency_check.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-security-audit/scripts/full_scan.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-security-audit/scripts/secrets_scan.py
- Executable scriptline 1
#!/usr/bin/env python3 - Network accessline 62
SecretPattern( name="Slack Webhook", pattern=re.compile(r'https://hooks\.slack\.com/services/T[a-zA-Z0-9_]+/B[a-zA-Z0-9_]+/[a-zA-Z0-9_]+'), - Network accessline 63
name="Slack Webhook", pattern=re.compile(r'https://hooks\.slack\.com/services/T[a-zA-Z0-9_]+/B[a-zA-Z0-9_]+/[a-zA-Z0-9_]+'), severity="MEDIUM", - Network accessline 64
pattern=re.compile(r'https://hooks\.slack\.com/services/T[a-zA-Z0-9_]+/B[a-zA-Z0-9_]+/[a-zA-Z0-9_]+'), - Network accessline 194
[Credential-related source omitted] - Network accessline 195
[Credential-related source omitted] - Network accessline 196
[Credential-related source omitted] - Access to secretsline 238
[Credential-related source omitted] - Access to secretsline 379
[Credential-related source omitted]
skills/Geek-skills-threejs-performance/references/assets.md
skills/Geek-skills-wechat-article-writer/assets/emoji-library.txt
- Possible obfuscationline 60
↗<U+FE0F> 右上箭头 | ↘<U+FE0F> 右下箭头 | ↩<U+FE0F> 返回箭头 | ⤴<U+FE0F> 向上弯箭头
skills/Geek-skills-wechat-article-writer/schemas/image-manifest.schema.json
skills/Geek-skills-wechat-article-writer/scripts/render_wechat_layout.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/Geek-skills-wechat-article-writer/scripts/validate_image_manifest.py
- Executable scriptline 1
#!/usr/bin/env python3
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Declared and executed: 2 facts
- Contacts hooks, which SKILL.md, the README and the plugin manifest do not mentionGeek-skills-security-audit/scripts/secrets_scan.py:64High confidence
- Contacts discord, which SKILL.md, the README and the plugin manifest do not mentionGeek-skills-security-audit/scripts/secrets_scan.py:196High confidence
Signal confidence: 184 facts
- Executable script in code of a file that runsGeek-skills-a-share-analyst/scripts/fetch_market_data.py:1High confidence
- Executable script in code of a file that runsGeek-skills-a-share-analyst/scripts/generate_report.py:1High confidence
- Executable script in code of a file that runsGeek-skills-a-share-analyst/scripts/strategy_multi_factor.py:1High confidence
- Executable script in code of a file that runsGeek-skills-a-share-analyst/scripts/technical_analysis.py:1High confidence
- Executable script in code of a file that runsGeek-skills-gaokao-expert/scripts/analyze_paper.py:1High confidence
- Executable script in code of a file that runsGeek-skills-gaokao-expert/scripts/analyze_question.py:1High confidence
- Access to secrets in code of the SKILL.mdGeek-skills-podcast-generator/SKILL.md:36High confidence
- Access to secrets in code of the SKILL.md: a variable, placeholder or call, not a stored valueGeek-skills-podcast-generator/SKILL.md:39Low confidence
- Network access in prose of documentationGeek-skills-podcast-generator/references/api_reference.md:17Low confidence
- Executable script in code of a file that runsGeek-skills-podcast-generator/scripts/generate_podcast.py:1High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueGeek-skills-podcast-generator/scripts/generate_podcast.py:46Low confidence
- Access to secrets in the frontmatter of the SKILL.mdGeek-skills-seedream-imagegen/SKILL.md:4Low confidence
- Access to secrets in prose of the SKILL.mdGeek-skills-seedream-imagegen/SKILL.md:22Low confidence
- Access to secrets in code of the SKILL.mdGeek-skills-seedream-imagegen/SKILL.md:53High confidence
- Access to secrets in prose of the SKILL.mdGeek-skills-seedream-imagegen/SKILL.md:66Low confidence
- Network access in prose of documentationGeek-skills-seedream-imagegen/references/api_reference.md:7Low confidence
- Network access in prose of documentationGeek-skills-seedream-imagegen/references/quick_start.md:11Low confidence
- Network access in prose of documentationGeek-skills-seedream-imagegen/references/quick_start.md:12Low confidence
- Access to secrets in code of documentationGeek-skills-seedream-imagegen/references/quick_start.md:20Low confidence
- Access to secrets in code of documentationGeek-skills-seedream-imagegen/references/quick_start.md:41Low confidence
- Access to secrets in code of documentation: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/references/quick_start.md:43Low confidence
- Access to secrets in code of documentationGeek-skills-seedream-imagegen/references/quick_start.md:48Low confidence
- Access to secrets in code of documentation: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/references/quick_start.md:52Low confidence
- Access to secrets in code of documentationGeek-skills-seedream-imagegen/references/quick_start.md:57Low confidence
- Access to secrets in code of documentation: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/references/quick_start.md:60Low confidence
- Access to secrets in code of documentation: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/references/quick_start.md:69Low confidence
- Executable script in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:1High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/scripts/generate_image.py:21Low confidence
- Access to secrets in prose of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:26Low confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/scripts/generate_image.py:28Low confidence
- Access to secrets in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:29High confidence
- Access to secrets in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:31High confidence
- Network access in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:34High confidence
- Access to secrets in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:36High confidence
- Network access in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:99High confidence
- Network access in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:100High confidence
- Network access in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:175High confidence
- Network access in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:176High confidence
- Network access in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:177High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/scripts/generate_image.py:202Low confidence
- Access to secrets in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:203High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/scripts/generate_image.py:204Low confidence
- Access to secrets in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:205High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/scripts/generate_image.py:209Low confidence
- Access to secrets in code of a file that runsGeek-skills-seedream-imagegen/scripts/generate_image.py:307High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueGeek-skills-seedream-imagegen/scripts/generate_image.py:362Low confidence
- Executable script in code of a file that runsGeek-skills-university-exam-prep/scripts/analyze_material.py:1High confidence
- Executable script in code of a file that runsGeek-skills-university-exam-prep/scripts/generate_questions.py:1High confidence
- Network access in prose of the SKILL.mdllm-wiki/SKILL.md:10Low confidence
- Executable script in code of a file that runsscripts/run_routing_evals.py:1High confidence
and 134 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md lab/Geek-skills-a-share-analyst/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Native | Manifest AGENTS.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md lab/Geek-skills-a-share-analyst/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md lab/Geek-skills-a-share-analyst/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md lab/Geek-skills-a-share-analyst/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md lab/Geek-skills-a-share-analyst/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.