Back to directory

LLM Wiki

An open-source template repository for building an LLM-maintained knowledge base: raw sources go in one folder, and an LLM ingests them into cross-linked wiki pages, answers questions by citing those pages, and audits the wiki for problems on request.

LLM Wiki

LLM Wiki follows Andrej Karpathy's LLM Wiki pattern in three layers: a raw folder holding immutable source material such as articles, transcripts or notes; a wiki folder of pages the LLM writes and maintains, including summaries, concepts, entities and syntheses; and a CLAUDE.md file that defines the schema telling the LLM how to structure everything for a given domain.

Three operations drive the workflow. Ingesting a source has the LLM create a summary page, create or update concept and entity pages, add cross-links, and update the index; asking a question has it search the wiki and answer with citations, optionally creating a synthesis page for a novel insight; and running a lint has it audit every page for orphans, contradictions, missing links and low-confidence claims, fixing what it can and reporting the rest. An optional Dataview dashboard surfaces low-confidence pages, recent updates and concepts by tag.

More in Agent memory

See all 450
  • A long technical guide to AI agent memory: why stateless LLMs need it, the different memory architectures from vector stores to knowledge graphs, and how Cognee implements one as an open-source memory layer.

  • A local, file-based long-term memory layer for a personal AI that never summarizes conversations, keeping every line with a timestamp and searching by time first and words second, rather than retrieving only similar-seeming chunks.

  • NEOTH

    deepwiki.com

    A local-first personal AI daemon in Rust with five memory tiers, consent-gated tools, a WASM plugin sandbox, multi-provider model routing and signed audit logs, aimed at people who want both a daily assistant and a serious coding helper.