Karpathy's LLM Wiki + Claude Code Boosts Coding Agents

Build a self-maintaining knowledge base in Obsidian using Karpathy's LLM Wiki blueprint and Claude Code: feed raw notes/docs into raw/ folder, auto-generate structured wiki/ markdown, query for precise code gen that improves via periodic linting.

LLM Wiki's Three-Layer Architecture Outperforms Static RAG

Karpathy's LLM Wiki creates a persistent, agent-navigable knowledge base superior to basic RAG by handling maintenance automatically. It uses three layers: (1) raw/ folder stores untouched source files like articles, notes, code snippets, screenshots, Figma links, HTML/CSS—your single source of truth; (2) wiki/ generates structured Markdown files with summaries, entities, interlinks, and an index.md for navigation; (3) schema rules dictate organization, updates, consistency checks, and cross-referencing.

Agents like Claude Code point to index.md, drill into relevant pages for context, reducing hallucinations and token waste. Humans explore ideas; LLMs manage tedious linking and upkeep, turning scattered notes into a connected base. Example: Farza Pedia processed 2,500 personal entries (diary, Apple Notes, messages) into hundreds of structured articles on friends, ideas, inspirations—built for agents to pull context for tasks like designing landing pages from past experiences.

Benefits include solving agent memory limits, enabling complex queries (e.g., "build CRM dashboard using referenced Chart.js charts"), and continuous improvement without manual edits. It's 10x more effective than RAG because the wiki self-evolves, spotting contradictions, stale info, missing links, and new connections via a "lint" prompt.

Quick Setup in Obsidian with Claude Code (Under 5 Minutes)

Install Obsidian (visualizes vaults, graph view for links) and Claude Code (e.g., in VS Code). Create a new Obsidian vault directory.

  1. Open Claude Code in the vault.
  2. Copy Karpathy's Gist (https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f#llm-wiki) into a file like llm-wiki.md.
  3. Paste this enhanced prompt into Claude Code: "Build me a complete LLM Wiki system based on this idea from Karpathy. I use Obsidian. Create the folder structure, initial scripts/tools if needed, and give me clear step-by-step instructions on how to ingest data and have you maintain the wiki. Make it practical and ready to run today."

Claude auto-creates raw/ and wiki/ folders, index.md, schema rules, and ingestion scripts. Tailor via description, e.g., "focus on frontend designs, UI inspirations, landing pages, design systems."

Use Obsidian Web Clipper browser extension to dump web content (markdown + images) directly into raw/. No custom code needed—leverages existing tools.

Ingest, Query, and Self-Improve for Production Coding

Ingest data: Drop files into raw/ (e.g., Tailwind docs, color/font notes, screenshots). Prompt Claude: "Compile new raw files into wiki: create summaries, extract concepts, add backlinks to index.md."

Query for code: Agents reference index.md for outputs. Example: Provided raw/ with Chart.js screenshots/HTML, UI snippets; prompted for CRM dashboard—generated app pulling exact components via cross-links, avoiding lazy model behavior or hallucinations.

Self-evolve: Run lint prompt: "Review entire wiki for contradictions, stale info, missing links, new connections. Fix and improve it." LLMs self-review prior work, enrich summaries, resolve issues—runs periodically as you add data. Feed more raw data and lint often for compounding accuracy.

Saves time on specialization, token costs (local images vs. scraping), enables specialized agents (e.g., frontend-focused). Graph view visualizes connections. Works with any coding agent; Claude Code integrates seamlessly for end-to-end workflows.

Video description
In this video, I show how Andrej Karpathy’s LLM Wiki — a self-evolving knowledge system — can be hooked up to Claude Code to massively supercharge your AI coding workflows. Prompt 1: Build me a complete LLM Wiki system based on this idea from Karpathy. I use Obsidian. Create the folder structure, initial scripts/tools if needed, and give me clear step-by-step instructions on how to ingest data and have you maintain the wiki. Make it practical and ready to run today. Prompt 2: Review the entire wiki for contradictions, stale info, missing links, or new connections. Fix and improve it. 🔗 My Links: Sponsor a Video or Do a Demo of Your Product, Contact me: intheworldzofai@gmail.com 🔥 Become a Patron (Private Discord): https://patreon.com/WorldofAi 🧠 Follow me on Twitter: https://twitter.com/intheworldofai 🚨 Subscribe To The SECOND Channel: https://www.youtube.com/@UCYwLV1gDwzGbg7jXQ52bVnQ 👩🏻‍🏫 Learn to code with Scrimba – from fullstack to AI https://scrimba.com/?via=worldofai (20% OFF) 🚨 Subscribe To The FREE AI Newsletter For Regular AI Updates: https://intheworldofai.com/ 👾 Join the World of AI Discord! : https://discord.gg/NPf8FCn4cD Something coming soon :) https://www.skool.com/worldofai-automation [Must Watch]: Claude Code Computer Use Can Control Your ENTIRE Computer! Automate Your Life!: https://youtu.be/KiywNP4b0aw?si=HuJnvik0AgLjIkCb Turn Antigravity Into AN AI Autonomous Engineering Team! Automate Your Code with Subagents!: https://www.youtube.com/watch?v=yuaBPLNdNSU Gemini 3.5? NEW Gemini Stealth Model Is POWERFUL & Fast! (Fully Tested): https://youtu.be/1abLcL33eKA?si=H50xRhJxVYM7HFPK 📌 LINKS & RESOURCES Github Repo: https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f#llm-wiki Claude Code: https://code.claude.com/docs/en/overview Obsidian: https://obsidian.md/ Obsidian Web Clipper: https://obsidian.md/clipper https://x.com/FarzaTV/status/2040563939797504467 https://x.com/karpathy/status/2040470801506541998 https://x.com/karpathy/status/2039805659525644595 Instead of manually writing notes or organizing data, the LLM builds and maintains a persistent wiki for you, constantly updating, cross-referencing, and improving itself. When paired with Claude Code, it can: Read your raw source files (articles, docs, code snippets) Maintain a fully structured markdown wiki Answer complex coding queries using your knowledge base Continuously improve over time, making AI-assisted coding smarter I’ll show you how to set it up in under 5 minutes, including the folder structure (raw/ and wiki/) and how to feed it your own data. If you want next-level AI coding, this is a game-changer. Features / Bullet Points ✅ LLM Wiki overview and how it works ✅ Connecting Claude Code to your self-evolving knowledge system ✅ Example: ingesting frontend designs and generating code suggestions ✅ How to query the wiki for smarter code outputs ✅ Why this approach is 10x more effective than RAG [Time Stamps]: 0:00 - Introductions 0:45 - Demo 1:45 - LLM Wiki Explanation 4:41 - Setup 8:16 - Frontend Example 11:19 - Output 12:07 - Enable Self-Eolving Tags / Keywords Claude Code, LLM Wiki, Andrej Karpathy, AI coding assistant, AI agent, self-evolving AI, AI knowledge base, LLM knowledge management, AI automation, Obsidian, markdown wiki, AI productivity, AI code generation, AI programming, Karpathy LLM, AI research tools Hashtags #ClaudeCode #LLMWiki #Karpathy #AIAssistant #AICoding #AIProductivity #SelfEvolvingAI #MarkdownWiki #Obsidian #AIWorkflow

Summarized by x-ai/grok-4.1-fast via openrouter

7517 input / 1504 output tokens in 9469ms

© 2026 Edge