prompt-hub

An MIT-licensed hub for high-quality LLM prompts and AI learning resources.

Structure

• docs/prompts/ - live prompts by category
• docs/learning/ - reading lists, notes, cheatsheets, roadmaps, glossary
• templates/ - shared templates for learning content
• CHANGELOG.md - history of changes
• LICENSE - MIT license

Getting started

• Prompts: browse docs/prompts/ (e.g., docs/prompts/prompt-booster.md).
• Learning: start at docs/learning/README.md — jump to the LLM Systems reading list or the LLM Roadmap.

Index generation

• Build topic index: powershell -ExecutionPolicy Bypass -File scripts/generate-topic-index.ps1
• Output file: docs/learning/topics.md

Docs site

• Preview locally: pip install mkdocs mkdocs-material then mkdocs serve (opens http://127.0.0.1:8000)
• Build static site: mkdocs build (outputs to site/)
• Optional deploy to GitHub Pages: mkdocs gh-deploy
• Config file: mkdocs.yml (sources are under docs/)

Contributing

Prompts

1. Add or update a .md in prompts/<category>/.
2. Keep titles clear and include tags.
3. Update CHANGELOG.md.

Learning content

1. Choose a type: learning/(reading-lists|notes|cheatsheets|roadmaps|papers|courses).
2. Start from a template in templates/ and include YAML front matter (title, summary, type, level, topics, updated).
3. Prefer concise commentary over raw link dumps.
4. Update learning/README.md if you add a new major topic.