AI Briefingllm_wiki: LLM Knowledge Management Built as a Wiki Instead of Document Reading
nashsu/llm_wiki
About the project
When you input materials in various formats such as PDFs, Office documents, and web clips, the LLM analyzes them and automatically converts them into interconnected wiki pages. Unlike traditional RAG methods that search the original sources every time to find answers, this structure organizes knowledge once, stores it, and keeps it up to date.
Two-stage Chain-of-Thought processing separates source analysis from wiki generation to improve accuracy. A knowledge graph based on four signals and Louvain community detection allow you to visually identify hidden connections between documents and knowledge gaps.
An agent running on a Rust-based backend integrates wikis, sources, graphs, and web search to provide answers. The Deep Research feature, utilizing Tavily or SerpApi, can collect additional information and automatically reflect it in the wiki.
This project concretizes Karpathy's LLM Wiki pattern as a desktop app. It maintains Obsidian compatibility and includes a built-in MCP server and HTTP API, making it easy to integrate with external AI agents like Claude Code to leverage knowledge.
nashsu/llm_wiki
LLM Wiki is a cross-platform desktop application that turns your documents into an organized, interlinked knowledge base — automatically. Instead of traditional RAG (retrieve-and-answer from scratch every time), the LLM incrementally builds and maintains a persistent wiki from your sources。
TypeScript
This introduction was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report errors, attribution issues, or removal requests via Contact.