Skip to content
repository radar
Agents & Orchestration

📚book-to-skill

ACTIVE BREAKOUT

virgiliojr94/book-to-skill

Turn any technical book PDF into a Claude Code skill, ready to study, reference, and use while you work.

⭐ Very popular: 22k stars, gaining about 3k a week

View on GitHub ↗

repo profile

vintage 2026 launched this year
language Python
license MIT

momentum

total stars 22k
stars added last week +3k
commits / week 15 accelerating · +300% vs prior mo
issues closed 90% ~2d to close, median
contributors 24 (+7 in 15d)
release cadence fortnightly
last activity 2d ago

durability

backing community / independent
openness permissive
bus factor 2 concentrated
top-author share 42% 6 mo

bus factor = how many people it takes to cover more than half the commits (6 months). 1 is a solo project; higher means the work is spread across a team. top-author share is the single busiest author's slice of those commits.

since we covered it

monthly average + 15k/mo (+116%/mo) · + 9k total since PR#39

why it's a big deal

  • It compiles knowledge instead of retrieving it. One extraction pass distills a book into named frameworks and decision rules; the maintainer’s measured “discovery loop tax” puts it at 24-51x fewer tokens than dumping the book into context per question.
  • It targets the open Skills spec as a compile target, so your library becomes portable infrastructure rather than vendor-locked embeddings - the clearest signal yet that the standard is real.
  • It goes well beyond books: fold architecture decision records, runbooks, brand guidelines or a cluster of papers into a single skill and keep updating it as new material lands. Internal knowledge becomes an installable capability.

under the hood

  • Pure-Python extraction with tiered parsers per format: pdftotext, pypdf or docling for PDFs (docling preserves tables and code blocks), ebooklib for EPUB, Calibre for MOBI; plain text and Markdown need no deps.
  • Output is a roughly 4,000-token SKILL.md core plus 1,000-token chapters; converting a full book costs about a dollar of model spend at current Sonnet pricing.
  • Ships its own measurement and validation tooling (discovery_tax.py, per-host skill validator); MIT-licensed converter that includes zero book content - you bring your own copy.

our take from PR#39, 2026-07-30

star history

PR#39 · 13k22k now May 2026Aug 2026
  1. PR#39 13k 2026-07-30
  2. now 22k + 9k since first covered

curve is sampled from GitHub's star history, plus our own daily readings since we covered it; the dashed stretch is before we first covered it, the solid line since. figures at coverage are the numbers we printed then (approx.), current count is live.

understory

Better known than its recent output, coasting a little on attention.

-20 understory score output 66 · clout 86
Aug 2025 Jul 2026
  • output, commits & releases
  • clout, star velocity

output = commits & releases; clout = star velocity, both 0 to 100 monthly indices; the gap where output runs above clout is the understory. The understory →

covered in

  • PR#39 2026-07-30 on the radar

    compile a book into an agent skill

similar projects

compare these →
  • hermes-agent

    11× the stars

    The agent that grows with you

    231k ACTIVE
  • 🖼️ Langflow

    7× the stars

    Langflow is a powerful tool for building and deploying AI-powered agents and workflows.

    153k ACTIVE
  • 📊 TradingAgents

    4.5× the stars

    TradingAgents: Multi-Agents LLM Financial Trading Framework

    98k ACTIVE

comments

Sign in with GitHub to add your blip on book-to-skill.

loading comments…