[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-75158":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":9,"languages":9,"totalLinesOfCode":9,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":13,"stars7d":15,"stars30d":16,"stars90d":14,"forks30d":14,"starsTrendScore":17,"compositeScore":18,"rankGlobal":9,"rankLanguage":9,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":20,"topics":23,"createdAt":9,"pushedAt":9,"updatedAt":24,"readmeContent":25,"aiSummary":26,"trendingCount":14,"starSnapshotCount":14,"syncStatus":27,"lastSyncTime":28,"discoverSource":29},75158,"skill-from-masters","GBSOSS\u002Fskill-from-masters","GBSOSS","Stand on the shoulders of giants — Create AI skills built on proven methodologies from domain experts.",null,1419,145,5,1,0,7,23,3,19.49,"MIT License",false,"main",true,[],"2026-06-12 02:03:33","# Skill From Masters\n\n> **Stand on the shoulders of giants** — Create AI skills built on proven methodologies from domain experts.\n\nA skill that helps you discover and incorporate frameworks, principles, and best practices from recognized masters before generating any new skill. Works with Claude Code, Codex, and other AI agent platforms.\n\n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT)\n\n---\n\n## Why This Skill?\n\n**The hard part of creating a skill isn't the format — it's knowing the best way to do the thing.**\n\nMost professional domains have masters who spent decades figuring out what works:\n- Jobs on product, hiring, and marketing\n- Bezos on writing (6-pager) and decision-making\n- Munger on mental models\n- Chris Voss on negotiation\n\nThis skill surfaces their methodologies before you write a single line, so your skill embodies world-class expertise from day one.\n\n## How It Works\n\n```\n1. You: \"I want to create a skill for user interviews\"\n\n2. Skill-from-masters:\n   ├── Checks local methodology database\n   ├── Searches web for additional experts\n   ├── Finds golden examples of great outputs\n   ├── Identifies common mistakes to avoid\n   └── Cross-validates across sources\n\n3. Surfaces experts:\n   - Rob Fitzpatrick (The Mom Test)\n   - Steve Portigal (Interviewing Users)\n   - Nielsen Norman Group best practices\n\n4. You select which methodologies to incorporate\n\n5. Extracts actionable principles from primary sources\n\n6. Hands off to skill-creator to generate the final skill\n```\n\n## Key Features\n\n| Feature | Description |\n|---------|-------------|\n| **3-Layer Search** | Local database → Web search for experts → Deep dive on primary sources |\n| **Golden Examples** | Finds exemplary outputs to define quality bar |\n| **Anti-Patterns** | Searches for common mistakes to encode \"don't do this\" |\n| **Cross-Validation** | Compares multiple experts to find consensus and flag disagreements |\n| **Quality Checklist** | Verifies completeness before generating |\n\n## Methodology Database\n\nThe skill includes a curated database covering 15+ domains:\n\n| Domain | Example Experts |\n|--------|-----------------|\n| Writing | Barbara Minto, William Zinsser, Amazon 6-pager |\n| Product | Marty Cagan, Teresa Torres, Gibson Biddle |\n| Sales | Neil Rackham (SPIN), Challenger Sale, MEDDIC |\n| Hiring | Laszlo Bock, Geoff Smart, Lou Adler |\n| User Research | Rob Fitzpatrick, Steve Portigal, JTBD |\n| Engineering | Martin Fowler, Robert Martin, Kent Beck |\n| Leadership | Kim Scott, Ray Dalio, Andy Grove |\n| Negotiation | Chris Voss, Fisher & Ury |\n| Startups | Eric Ries, Paul Graham, YC |\n| Decision Making | Jeff Bezos, Charlie Munger, Annie Duke |\n\nPlus an **\"Oral Tradition\"** section for experts who share primarily through talks and interviews (Steve Jobs, Elon Musk, Jensen Huang, Patrick Collison, etc.)\n\n## Installation\n\n### Option 1: Clone directly (Recommended)\n\n```bash\n# Navigate to your skills directory\ncd ~\u002F.claude\u002Fskills  # Claude Code\n# or your platform's skills path\n\n# Clone this skill\ngit clone https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fskill-from-masters.git\n```\n\n### Option 2: Manual installation\n\n1. Download the `skill-from-masters` folder\n2. Place it in your skills directory\n\n### Dependency\n\nThis skill works with **skill-creator**. Make sure skill-creator is available in your environment.\n\n## Usage\n\nThe skill automatically activates when you ask to create a new skill. It runs **before** skill-creator to ensure your skill is built on proven methodologies.\n\n### Example Session\n\n**You:** \"Help me create a skill for writing product requirement documents\"\n\n**Skill-from-masters:**\n> Let me search for the best methodologies on PRD writing.\n>\n> *[Searches: \"PRD best practices expert\", \"product requirements document methodology\"]*\n>\n> I found several strong frameworks:\n>\n> **Product Management:**\n> - Marty Cagan: Problem-first PRDs, separating problem space from solution space\n> - Teresa Torres: Outcome-focused requirements\n>\n> **Business Writing:**\n> - Barbara Minto: Pyramid Principle - lead with conclusion, MECE structure\n> - Amazon: Working backwards, 6-pager narrative format\n>\n> **Golden Examples Found:**\n> - Intercom's public PRD template\n> - Linear's product spec format\n>\n> **Common Mistakes:**\n> - Jumping to solutions before defining the problem\n> - Writing for engineers instead of for alignment\n> - Missing success metrics\n>\n> Which approaches resonate? Or should I search deeper into any of them?\n\n**You:** \"I like Amazon's approach combined with Cagan's problem-first thinking.\"\n\n**Skill-from-masters:** *[Fetches primary sources on Amazon's methodology, extracts principles, then generates skill via skill-creator]*\n\n## Skills\n\nThis repository contains the following skills:\n\n| Skill | Description |\n|-------|-------------|\n| **[skill-from-masters](skill-from-masters\u002FSKILL.md)** | Create new skills based on proven methodologies from domain experts. |\n| **[search-skill](skills\u002Fsearch-skill\u002FSKILL.md)** | Search for existing skills from trusted marketplaces. |\n| **[skill-from-github](skills\u002Fskill-from-github\u002FSKILL.md)** | Learn from high-quality GitHub projects and create skills based on that knowledge. |\n\n### skill-from-masters\n\nWhen you want to **create a new skill based on expert methodologies**:\n\n- 3-layer search: local database → web experts → primary sources\n- Finds golden examples and anti-patterns\n- Cross-validates across multiple experts\n- Hands off to skill-creator for final generation\n\n**Example:**\n```\nYou: \"Help me create a skill for user interviews\"\n→ Finds: Rob Fitzpatrick (The Mom Test), Steve Portigal, Nielsen Norman Group\n→ You select which methodologies to incorporate\n→ Generates skill with those principles encoded\n```\n\n### search-skill\n\nWhen you want to **find an existing skill** instead of creating one:\n\n- Searches only 5 trusted sources (no random internet results)\n- Tier-based priority: official → curated → aggregators\n- Filters out low-quality results (stars \u003C 10, outdated, no SKILL.md)\n- Security checks for suspicious code patterns\n\n**Example:**\n```\nYou: \"I need a skill for frontend design, automated testing, and code review\"\n→ Searches: anthropics\u002Fskills, ComposioHQ, travisvn, skills.sh, skillsmp.com\n→ Returns: frontend-design (official), webapp-testing (official), code-review-excellence (26k stars)\n```\n\n### skill-from-github\n\nWhen you want to **learn from a GitHub project** and turn that knowledge into a skill:\n\n- Search GitHub for quality projects (stars > 100, actively maintained)\n- Present options and wait for your confirmation\n- Deep dive into selected project (README, source code, examples)\n- Summarize what it learned, then create skill via skill-creator\n\n**Example:**\n```\nYou: \"I want to convert images to ASCII art\"\n→ Searches GitHub, finds: ascii-image-converter (3.1k stars), RASCII (224 stars)\n→ You select ascii-image-converter\n→ Learns: brightness-to-character mapping, aspect ratio handling, color techniques\n→ Creates skill encoding that knowledge (not just wrapping the tool)\n```\n\n**Key difference:** This skill encodes the *knowledge* from projects, so the skill works even without the original tool installed.\n\n## File Structure\n\n```\nskill-from-masters\u002F\n├── skill-from-masters\u002F\n│   ├── SKILL.md                              # Core skill: create from expert methodologies\n│   └── references\u002F\n│       ├── methodology-database.md           # Curated expert frameworks\n│       └── skill-taxonomy.md                 # 11 skill type categories\n├── skills\u002F\n│   ├── search-skill\u002F\n│   │   └── SKILL.md                          # Search existing skills from trusted sources\n│   └── skill-from-github\u002F\n│       └── SKILL.md                          # Learn from GitHub projects\n├── README.md\n├── LICENSE\n└── .gitignore\n```\n\n## Quality Checklist\n\nBefore finalizing any skill, this skill verifies:\n\n- [ ] Searched beyond the local database\n- [ ] Found primary sources, not just summaries\n- [ ] Found golden examples of the output\n- [ ] Identified common mistakes to avoid\n- [ ] Cross-validated across multiple experts\n- [ ] Encoded specific, actionable steps (not vague principles)\n\n## Contributing\n\nContributions welcome! Especially:\n\n- Adding new domains and experts to the methodology database\n- Improving framework descriptions with source links\n- Sharing examples of skills created with this approach\n\nPlease:\n1. Fork the repository\n2. Create a feature branch\n3. Submit a pull request\n\n## License\n\nMIT License — feel free to use, modify, and distribute.\n\n---\n\n**Philosophy:** Quality isn't written. It's selected.\n","Skill From Masters 是一个帮助用户基于领域专家验证过的方法论创建AI技能的项目。它通过三层搜索机制（本地数据库、网络专家搜索、原始资料深入挖掘）来发现和整合最佳实践，同时提供黄金案例和反模式以确保高质量输出，并且能够跨多个来源进行交叉验证。此工具适合那些希望在开发新技能前就融入世界级专业知识的人士使用，无论是产品设计、销售策略还是用户研究等领域，都能从中受益。",2,"2026-06-11 03:52:33","high_star"]