abubakarsiddik31

Inspired by Andrej Karpathy's llm-wiki — the idea that LLMs should maintain a persistent, compounding wiki rather than re-derive answers from raw sources on every query. https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f

20
1
100% credibility
Found Apr 12, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

AI-powered personal knowledge base that automatically ingests documents into an interconnected wiki of markdown pages.

How It Works

1
🔍 Discover Axiom Wiki

You learn about a helpful tool that turns your documents into a smart, self-updating knowledge collection.

2
🛠️ Follow the setup guide

Answer a few simple questions to pick your thinking helper and choose folders for your notes and wiki.

3
📥 Add your documents

Drop notes, articles, PDFs, or code files into your source folder and feel the magic begin.

4
🤖 Process everything

Tell it to read your files, and it pulls out key people, ideas, and links to build connected pages.

5
💬 Ask away

Chat with your wiki to get clear answers backed by your own knowledge with easy links to explore more.

🌟 Enjoy your growing wiki

Your knowledge base updates itself as you add more, staying organized and insightful forever.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 20 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 axiom-wiki?

Axiom-wiki is a TypeScript CLI tool that builds and maintains a personal knowledge base as an interconnected Markdown wiki. Drop PDFs, docs, images, or codebases into a raw folder, and it uses LLMs like Gemini or Ollama to extract entities, concepts, and summaries into categorized pages—automatically linking them without re-deriving from scratch each time. Inspired by Andrej Karpathy's llm-wiki (https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f), it compounds knowledge persistently, outputting Obsidian-compatible vaults with frontmatter.

Why is it gaining traction?

Unlike basic RAG tools that query raw sources repeatedly, axiom-wiki compiles a living graph of pages, with incremental ingest skipping unchanged files via hashes and git-diff sync for codebases. Standout CLI commands like `map` (auto-plans project docs), `clip` (grabs web articles), `sync` (updates on changes), and `graph` (visualizes links) make it dead simple to use. Multi-LLM support, cost tracking, and MCP integration for tools like Cursor hook devs tired of manual notetaking.

Who should use this?

Solo developers mapping codebases for quick onboarding, researchers ingesting papers into a queryable wiki, or indie hackers clipping articles and syncing project notes. Ideal for those building axiom wiki-style bases around Minecraft mods, WWE wrestlers like Axiom, or LoL arc wiki entries—anywhere compounding answers from scattered sources beats starting over.

Verdict

Grab it via npm for local project wikis—early at 20 stars and 1.0% credibility, but polished docs, CI, and Vitest coverage make it production-ready for personal use. Worth the 2-minute setup if you want a self-maintaining knowledge graph without the fluff.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.