LLM Wiki Compiler CLI
llm-wiki-compiler
A knowledge compiler: turns raw sources into an interlinked wiki with answer verification and approval, with an MCP mode
- 2,223 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Reads and writes files and sends materials to the model provider
- On native Windows verification is not finished
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/llm-wiki-compilerDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Reads and writes files and sends materials to the model provider
- On native Windows verification is not finished
This is third-party code. Review the repository files before installing.
What it does
The tool turns raw materials into an interlinked wiki following Andrej Karpathy's LLM Wiki idea. Since version 1.4.0 answers are verified and approved before publication. There is an MCP mode: source intake, wiki building, queries, page search, a linter, evaluation and OKF package exchange. Installation: `npm install -g llm-wiki-compiler`, version 1.4.2 from 03.10.2026, MIT license, TypeScript.
Who it is for. For people who build a knowledge base from documents and articles and want an agent to maintain the wiki and answer from it with verification.
Good fit when
- You need to turn a set of documents into an interlinked wiki
- You need answers from the base to be verified and approved
- You need an MCP mode for an agent to work with the wiki
Not a fit when
- You need native Windows support with full verification: per the README it is not finished
- You cannot send materials to a model provider
Example request
Build a wiki from the folder of project notes, link the pages and show which answers await my approvalLimitations
The tool reads and writes files and calls a model, so the base's materials go to the chosen provider. Per the README on native Windows verification and CI are not finished, and file restriction guarantees are not verified. Answer approval since version 1.4.0 reduces the danger but does not replace your own check. Access is extended.
How to disable. Remove the package with npm uninstall -g llm-wiki-compiler and the wiki folder if it is no longer needed.
FAQ
What does the tool do?
It turns raw materials into an interlinked wiki and verifies answers before publication.
How do I install it?
npm install -g llm-wiki-compiler.
Is there an MCP mode?
Yes: source intake, building, queries, page search, a linter, evaluation and OKF package exchange.
Discussion
No questions or reviews yet.
Related
CLI toolsLLM Wiki
llm-wiki
Knowledge bases that AI compiles: topics, import, llm-wiki lint checks, the /wiki:ingest and /wiki:archive commands. A plugin for Claude Code and Codex
Turns session history of Claude Code, Codex, Cursor and Gemini CLI into a searchable wiki on Karpathy's LLM Wiki pattern, locally
Karpathy LLM Wiki
karpathy-llm-wiki
A skill for a personal knowledge base the model maintains: ingest sources, compile cited articles, lint integrity
A workflow for Claude Code, Codex and Gemini CLI where the agent files source documents into a persistent, linked wiki
An MCP server for Yandex Wiki: read pages by slug or URL, create, update and append, with a read-only mode
A skill set that lets an agent run a personal Obsidian wiki: it compiles knowledge from documents and past sessions and answers with citations
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| LLM Wiki Compiler (this entry) | no | 2,223 | |
| LLM Wiki | no | 1,479 | knowledge bases compiled by AI with llm-wiki lint checks |
| Second Brain Starter | no | 841 | a PRD and a phased plan for a personal second brain |
README in short
The README describes a knowledge compiler that builds a wiki from materials, with answer verification since version 1.4.0 and an MCP mode. Installation through npm. MIT license.