Agent skill · bighardperson
llm-wiki
Build and maintain a personal knowledge base (wiki) using LLMs. Instead of RAG-style retrieval, the LLM incrementally compiles, cross-references, and maintains a persistent structured wiki from raw sources. Use when user wants to create a knowledge base, build a personal wiki, organize research notes, ingest documents into a structured wiki, or maintain a living knowledge repository.
What it needs
About 4k tokens when loaded.
What this skill does
Karpathy LLM Wiki A pattern for building personal knowledge bases using LLMs. Based on Andrej Karpathy's LLM Wiki. The Core Idea Instead of RAG (retrieve-on-every-query), the LLM incrementally builds and maintains a persistent wiki — a structured, interlinked collection of markdown files. When you add a new source, the LLM reads it, extracts key information, and integrates it into the existing wiki — updating entity pages, revising topic summaries, noting contradictions, and strengthening the evolving synthesis. The wiki is a persistent, compounding artifact. Cross-references are already there. Contradictions have been flagged. Synthesis reflects everything you've read. It gets richer with every source and every question. You never write the wiki yourself — the LLM writes and maintains all of it. You're in charge of sourcing, exploration, and asking the right questions. Architecture (Three Layers) 1. Raw Sources Your curated collection of source documents — articles, papers, images, data files. These are immutable. The LLM reads from them but never modifies them. This is your source of truth. 2. The Wiki A directory of LLM-generated markdown files — summaries, entity pages, concept pages, comparisons, overview, synthesis. The LLM owns this layer entirely: creates pages, updates them when new sources arrive, maintains cross-references, keeps everything consistent. You read it; the LLM writes it. 3. The Schema A configuration document (this skill) that tells the LLM the wiki's structure, conventions, and workflows. You and the LLM co-evolve this over time. Operations Ingest User drops a new source. The LLM: 1. Reads the source and discusses key takeaways 2. Writes a summary page in the wiki 3. Updates the index 4. Updates relevant entity and concept pages across the wiki 5. Appends an entry to the log A single source may touch 10-15 wiki pages. Prefer ingesting one source at a time for quality. Query Ask questions against the wiki. The LLM: 1. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills bighardperson/Karpathy LLM Wiki