tlehman

tlehman / litprog-skill

Public

Literate programming skill for agent harnesses like Claude Code, OpenCode and Hermes Agent

57
3
100% credibility
Found Mar 11, 2026 at 47 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A Claude Code skill that converts existing codebases into literate programming documents blending narrative prose, diagrams, and code, supporting extraction of code files and automatic synchronization when editing sources.

How It Works

1
📚 Discover the tool

You hear about a helpful way to turn your code into an easy-to-read story with explanations and pictures.

2
🛠️ Add to your workspace

You place this special helper into your coding area so your AI assistant can use it.

3
💬 Ask AI to explain code

You tell your AI friend to review your code and create a storytelling document.

4
✨ Get narrative document

You receive a beautiful file mixing words, drawings, math, and highlighted code in the best order for understanding.

5
🖊️ Edit your code files

You tweak your regular code files as usual, and the story document updates itself to match.

6
📄 Make a pretty PDF

You turn the story into a polished, printable guide with a table of contents.

🎉 Share clear code stories

Now anyone can read and understand your project from start to finish, perfect for teams or learning.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 47 to 57 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is litprog-skill?

Litprog-skill is a TypeScript-based skill for agent harnesses like Claude Code, OpenCode, and Hermes Agent, turning regular codebases into literate programming markdown files. Inspired by literate programming Donald Knuth, it generates .lit.md docs that mix prose, Mermaid diagrams, and code chunks in psychological order—easy for humans, not compilers—then weaves them into polished PDFs via Pandoc or tangles them back to runnable source. Developers get a single narrative source of truth for code explanation and generation, solving the gap between scattered docs and actual implementation.

Why is it gaining traction?

It stands out by plugging directly into AI agents with a simple /literate-programming command, auto-analyzing code and spitting out literate programming examples complete with diagrams and math. The reverse-sync hook automatically merges direct source edits back into the .lit.md without manual hassle, keeping docs alive. Unlike static literate programming Emacs or Org-mode tools, this fits modern agent-driven workflows, producing literate programming PDFs that double as onboarding books.

Who should use this?

Backend engineers maintaining tangled monoliths who need to explain architectural why's to juniors. Teams using Claude Code or similar agents for literate programming Python or TypeScript projects, especially for code reviews and knowledge transfer. Solo devs prototyping systems where READMEs fall short, craving literate programming GitHub repos that weave/tangle effortlessly.

Verdict

Try it if you're in the Claude ecosystem and want literate programming Knuth-style without the org-mode learning curve—solid docs and CLI make it approachable despite 19 stars and 1.0% credibility score signaling early maturity. No tests yet, so verify tangles on toy projects first; high potential for agent-heavy teams.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.