book-to-skill

name	book-to-skill
description	Convert technical books and documents (PDF, EPUB, DOCX, HTML, Markdown, RTF, MOBI) into structured agent skills with frameworks, mental models, chapter references, and decision rules. Includes full extraction pipeline. Use when the user wants to turn a book or document collection into a reusable agent skill for study and reference.
domain	core
author	mahipal, virgiliojr94
license	MIT
subdomain	tooling
tags	["documentation","skill-generation","knowledge-management","pdf","epub","conversion","learning","extraction"]
version	1.0.0

persona: name: "Virgilio Jr." title: "The Knowledge Distiller — Book-to-Skill Pipeline Maintainer" expertise: ['Document Extraction', 'Knowledge Structuring', 'Agent Skill Design', 'Pipeline Engineering'] philosophy: "Books contain crystallized expertise. Extract structure, not summaries." credentials: ['Creator of book-to-skill (MIT)', 'Open-source maintainer'] principles: ['Structure over summary', 'Preserve author precision', 'Layer depth appropriately', 'Privacy-first: all processing local']

Book-to-Skill Converter

Transform written knowledge into actionable agent skills by extracting frameworks, principles, techniques, and anti-patterns — not producing summaries.

Upstream repo: github.com/virgiliojr94/book-to-skill
Upstream author: @virgiliojr94
This integration: Full extraction pipeline vendored into core/book-to-skill/
License: MIT
Format support: PDF, EPUB, DOCX, TXT, Markdown, reStructuredText, AsciiDoc, HTML, RTF, MOBI/AZW/AZW3

---

When to Use

Trigger phrases:

• "turn this book into a skill"
• "book-to-skill"
• "convert this PDF to an agent skill"
• "generate a skill from this document"
• "study this book with my agent"

Use when:

• You want your agent to reference a technical book while coding
• You have a collection of docs/papers you constantly re-read
• You need structured access to a book's content without dumping the whole PDF into context
• You own a book PDF and want it as an on-demand agent skill

Don't use for:

• Fiction / narrative books (the tool works technically but the output isn't useful)
• Books you don't own (copyright compliance — you must own the source)
• Simple web articles (use web extraction instead; full pipeline is overkill)

---

How It Works

PDF / EPUB / DOCX / HTML / Markdown / RTF / MOBI
    │
    ▼
Extract text → Detect structure (chapters, headings)
    │
    ▼
LLM analyzes: frameworks, principles, techniques, anti-patterns
    │
    ▼
Generate structured skill:
  ├── SKILL.md          (core mental models + chapter index, ~4K tokens)
  ├── chapters/ch*.md   (one per chapter, ~1K tokens each, loaded on-demand)
  ├── glossary.md       (key terms with chapter refs, ~1.5K tokens)
  ├── patterns.md       (techniques, algorithms, design patterns, ~2K tokens)
  └── cheatsheet.md     (decision tables, quick-reference rules, ~1K tokens)