Virgilio Junior · @virgiliojr94 ↗ book-to-skill
Turn any technical book or folder of docs into a structured agent skill your agent loads on demand: a compact core plus per-chapter files and a cheatsheet. It answers at a fraction of the tokens.
The roster
What it does
book-to-skill turns a technical book, a docs folder, or a stack of sources into a structured skill your agent loads on demand. Point it at a PDF, EPUB, or folder and it distills the material into a compact core file plus per-chapter files, a glossary, a patterns doc, and a cheatsheet, keeping the structure rather than flattening it into a summary.
Once installed, you ask about a topic and the agent reads the right chapter and answers from the real content, with no hallucination and no digging through a PDF. Answering one question costs 24 to 51 times fewer tokens than dumping the whole book into context. The same extraction works on internal docs, brand systems, and specs, and it runs in Claude Code, GitHub Copilot CLI, and Amp.
When to use it
- You want your agent to know a specific book or doc set
Turn any technical book or folder of docs into a skill it loads on demand.
- You need the knowledge without the token cost
A compact core plus per-chapter files answer at a fraction of the tokens.
Common questions
What file formats can it read?
book-to-skill turns PDF, EPUB, DOCX, Markdown, HTML, RTF, and MOBI sources into an agent skill, so a book or a folder of docs becomes a compact core plus per-chapter files and a cheatsheet.
Which agents does it work with?
book-to-skill produces a standard Agent Skill your agent loads on demand, and it works with GitHub Copilot CLI, Amp, and Claude Code.
Is book-to-skill free?
Yes. book-to-skill is open-source under the MIT license. A standalone pip install book-to-skill gives you just the text extractor if you want the engine on its own.