OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

LLM Wiki

LLM Wiki is an open-source personal knowledge-base application that indexes documents, notes and web clippings, and connects Claude to the resulting wiki through MCP. Its web app displays linked wiki pages and their sources, while a scheduled Claude Routine can synthesize new material into updated pages.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

LLM Wiki — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use LLM Wiki when you want to turn a growing collection of documents, notes and annotated reading into a connected reference that Claude can search and maintain.

Who it is for

  • People who want a searchable, evolving reference built from their reading and research.
  • Claude users who want an MCP-connected assistant to work with their own source material.
  • Teams exploring a shared institutional knowledge base from accumulated documents and notes.

Documented uses

  • Index a folder of research documents for browsing and AI access.
  • Capture webpages and PDFs with highlights and margin notes.
  • Use Claude to create or update wiki pages based on accumulated sources.
  • Navigate links between concepts, pages and their underlying sources.
Before you choose

Requirements & limits

The repository says the project can be self-hosted or tried at llmwiki.app. Local setup requires Python 3.11+ and Node.js 20+.

  • Claude access is provided through MCP, with setup instructions for Claude Desktop and Claude Code.
  • A Chrome extension captures webpages and PDFs, including highlights and comments.
  • The web app supports source viewing, page cross-links, a graph viewer and visualizations.
  • Local mode is loopback-only and does not support LAN or remote binding.
  • The repository is licensed under Apache-2.0.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a local LLM Wiki workspace

Install the project, index a folder of research files and open the local web app.

Before you start

  • Python 3.11 or later
  • Node.js 20 or later
  • A local folder for PDFs, Word documents, PowerPoints, Markdown files or notes

Steps

  1. Clone the repository and install the Python requirements for the API and MCP server, then install the web dependencies with npm.

  2. Run the CLI's open command with the path to your research folder.

  3. Open the local web app at localhost:3000.

Expected outcome

The folder is indexed and the local API and web app start, making its files available in the workspace.

Guide limitations

  • Local mode binds the API to 127.0.0.1 and does not support LAN or remote binding.
  • LibreOffice is optional for Word and PowerPoint extraction; MISTRAL_API_KEY is optional for higher-quality PDF OCR.
Documentation 1 ↗
Source-based practical guide

Connect a local workspace to Claude over MCP

Add an initialized workspace as an MCP server entry for Claude Desktop or Claude Code.

Before you start

  • An initialized local LLM Wiki workspace
  • Claude Desktop or Claude Code

Steps

  1. Run the CLI's mcp-config command with the workspace folder.

  2. Add the printed JSON to claude_desktop_config.json for Claude Desktop or .claude/settings.json for Claude Code.

  3. Add a separate MCP server entry for each workspace folder.

Expected outcome

Claude can read, write and search the connected wiki through MCP.

Guide limitations

  • The guide documents Claude Desktop and Claude Code configuration; the exact setup for other MCP-compatible apps is not included here.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
8 October 2026
Original website title
lucasastorian/llmwiki · Official homepage title · retained in its source language
How we review sourcesSuggest a correction