Skip to content

Karpathy's LLM Wiki pattern

A reference summary of the personal knowledge base maintained by an LLM pattern, as articulated by Andrej Karpathy in late 2025. This page is documentation, not a skill — it captures the idea so we can decide later whether to vendor an existing implementation (e.g. obsidian-second-brain) or author a local skill of our own.

Sources

TL;DR

The wiki is a persistent, compounding artifact. RAG re-derives knowledge from raw sources on every query; an LLM Wiki compiles that knowledge once into structured markdown and keeps it current as new sources arrive.

Obsidian is the IDE; the LLM is the programmer; the wiki is the codebase. — Karpathy, LLM Wiki gist

The original tweet (verbatim)

LLM Knowledge Bases

Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So:

Data ingest: I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them.

IDE: I use Obsidian as the IDE "frontend" where I can view the raw data, the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides).

Q&A: Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale.

Output: Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base.

Linting: I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into.

Extra tools: I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries.

Further explorations: As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows.

TL;DR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

Architecture

Three layers, from the gist:

  1. Raw sources — your curated source documents (articles, papers, images, datasets). Immutable — the LLM reads but never modifies them. This is your source of truth.
  2. The wiki — a directory of LLM-generated markdown (summaries, entity pages, concept pages, comparisons, an overview, a synthesis). The LLM owns this layer entirely; you read it, the LLM writes it.
  3. The schema — a CLAUDE.md / AGENTS.md that tells the LLM how the wiki is structured, what the conventions are, and what workflows to follow. You and the LLM co-evolve this over time.

Operations

Operation What it does
Ingest Drop a new source into raw/. The LLM reads it, summarizes, files it under the right entity/concept pages, updates the index, appends to the log. A single source might touch 10–15 pages.
Query Ask a question. The LLM reads the index, drills into relevant pages, and answers — as markdown, a comparison table, a Marp deck, a matplotlib chart. File the good answers back into the wiki so explorations compound.
Lint Periodic health check. Find contradictions, stale claims, orphan pages, missing cross-references, data gaps. Suggests new questions to investigate and new sources to look for.

Two special files

  • index.md — content-oriented catalog. Every page listed with a link and one-line summary, organized by category. The LLM updates it on every ingest and reads it first when answering a query. At ~100 sources / a few hundred pages this works without embedding-based RAG.
  • log.md — chronological, append-only. Use a consistent prefix (e.g. ## [2026-04-02] ingest | Article Title) so the log becomes greppable: grep "^## \[" log.md | tail -5.

Why this works

Maintenance — not reading or thinking — is what kills personal wikis. Updating cross-references, keeping summaries current, noting contradictions across dozens of pages: humans give up because the burden grows faster than the value. LLMs don't get bored, don't forget a back-link, and can touch 15 files in one pass. The wiki stays maintained because the cost of maintenance is near zero.

The pattern is, in spirit, Vannevar Bush's Memex (1945) — a private, curated knowledge store with associative trails. The part Bush couldn't solve was who does the maintenance. The LLM does.

Ecosystem

Implementations and adjacent tools that have grown around this pattern:

  • eugeniughelbur/obsidian-second-brain — a Claude Code skill (not an Obsidian plugin) that pushes the pattern further: 31 slash commands + scheduled agents + Python scripts. Adds AI-first conventions (machine-readable preambles, bi-temporal facts, a "Two-Output Rule" where every answer also updates relevant pages), vault-first research that scans existing notes before going to the web, and _CLAUDE.md / index.md / log.md / SOUL.md / CRITICAL_FACTS.md (≈120 tokens always loaded) at the vault root. "If Karpathy's wiki is a knowledge base you maintain with an LLM, this is a knowledge base that maintains itself."
  • tobi/qmd — local search engine for markdown (hybrid BM25 + vector + LLM re-ranking, on-device). Has both a CLI for the LLM to shell out to and an MCP server. Mentioned by the gist as the natural upgrade path once the index file alone stops scaling.
  • Obsidian Web Clipper, Marp (markdown-to-slides), Dataview (frontmatter queries) — the supporting tools the gist recommends.

Relation to this repo

This repo's project-knowledge-harness is task memory — TODO.md + backlog/ + pitfalls/, oriented around shipped/deferred work and past traps. Karpathy's LLM Wiki is knowledge memory — a curated, synthesized view of an external research area. The two are orthogonal and could coexist on the same project.

Possible follow-ups (not committed):

  • Vendor obsidian-second-brain into skills/vendor/ (license permitting), alongside project-knowledge-harness as a "two flavors of memory" pair.
  • Author a minimal local llm-wiki-bootstrap skill — just the gist's three-layer architecture + index.md / log.md + ingest / query / lint commands, no Obsidian dependency.

If you want to track either, add a P3 entry via scripts/add-todo.sh.