Zimi MCP Server
Allows managing and searching offline ZIM archives from Kiwix, including downloading from the Kiwix catalog, full-text search, and reading articles.
Provides tools to search and read Wikipedia articles from offline ZIM files, with support for language switching, cross-article navigation, and autocomplete.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Zimi MCP Serversearch for astrophysics articles"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Zimi
A modern experience for your ZIM files.
Kiwix packages the world's knowledge into ZIM files. Zimi makes them feel like the real internet with a rich web UI, fast JSON API, and an MCP server for AI agents. Everything works offline, in your language.
What it does
Search everything. One query across every ZIM, 100M+ articles, with
-word,"exact words",OR,lang:andin:when you need them.Apps over your library. Zimipedia for every wiki, Bookshelf for every book, ZimiTube for video, ZimiExchange for Q&A, Reddot for subreddits, and offline street Maps.
Keep what you find. Bookmarks, lists, likes and highlights in every app, synced across devices when you sign in.
A real library. The whole Kiwix catalog one click away, even offline, with auto-updates and BitTorrent downloads that seed back.
Make your own ZIMs. A web page, a whole site, a video playlist, a subreddit, a web archive or a folder of your files.
Share on your network. Zimi devices find each other and pass ZIMs at LAN speed.
Every language. Articles across languages by Wikidata ID; ten UI languages, right-to-left included.
For people and agents. Web app, JSON API and an MCP server. Runs on Docker, Podman, pip, macOS, Windows, Linux, or your phone as a PWA.
Related MCP server: ZIM RAG MCP Server
Screenshots
Homepage | Search Results |
|
|
Zimipedia | Bookshelf |
|
|
Maps | ZimiTube |
|
|
Reddot | Catalog |
|
|
Sharing | Make a ZIM |
|
|
Install
macOS
brew tap epheterson/zimi
brew trust --tap epheterson/zimi # Homebrew 6 and later
brew install --cask zimiLinux: sudo snap install zimi, or the AppImage (it updates itself; the window needs WebKitGTK 4.1 from your distro, otherwise Zimi opens in your browser).
Windows: the installer or portable zip from Releases.
Docker
docker run --network host -v ./zims:/zims -v ./zimi-config:/config epheterson/zimi/zims holds your ZIM files, /config keeps settings and indexes. Open http://localhost:8899. Host networking lets LAN discovery and seeding work; for bridge mode, Compose, rootless Podman and Kubernetes see operations and networking.
Python
pip install zimi
ZIM_DIR=./zims zimi serve --port 8899Two packages: pip install zimi is the complete experience; pip install zimi-mcp is only the MCP server and search, for AI agents (280 KB, no web app).
Most people configure nothing. Everything is in Settings; zimi config prints every effective value and where it came from, and each ZIMI_* variable is documented in the guides below.
Documentation
Reading: search, the reader, Reader View, find in page, languages, PDFs, offline use.
Apps: Zimipedia, Bookshelf, ZimiTube, ZimiExchange, Reddot and Maps.
Saving: bookmarks, lists, likes, highlights, sync.
Making ZIMs: every capture mode, folders and their metadata files, web archives.
Getting & sharing: the catalog, downloads, auto-update, BitTorrent, Nearby.
Access: public access, accounts, per-ZIM allowlists, single sign-on.
Operations: configuration, backup, air-gapped use, monitoring, updates.
API & MCP: the JSON API (
/openapi.json), RAG chunks, the MCP server and its tools.
MCP server
pip install zimi-mcp
zimi-mcp /path/to/zims{
"mcpServers": {
"zimi": { "command": "zimi-mcp", "args": ["/path/to/zims"] }
}
}zimi-mcp is the MCP server and search alone, built from this repository and released with it. pip install "zimi[mcp]" gives the full install the same command, and zimi mcp. Three tools by default (search, read a page as Markdown, read one section), about 360 tokens of schema, sized for small local models; --tools full serves every tool. No web server is started. For Docker on another machine, run docker exec -i zimi python3 -m zimi mcp over ssh. See API & MCP for the tools and measured numbers, and the SearXNG and OpenWebUI guides.
Contributing
Issues and ideas are welcome, and every one gets answered. See CONTRIBUTING.md; security reports go privately per SECURITY.md.
License
MIT. Desktop and Docker builds bundle libtorrent-rasterbar (BSD-3-Clause), see THIRD_PARTY_NOTICES.md.
Built with ❤️ in California by @epheterson and Claude Code.
This server cannot be deployed
Maintenance
Related MCP Connectors
Persistent memory and knowledge management for AI agents with semantic search and 50+ tools.
Persistent docs and memory for AI agents — read, write, organize & search a shared workspace.
15 media & data tools for AI agents: search, transcribe, subtitles, voiceover, translate & more.
Universal memory for AI agents and tools. Save, organize and search context anywhere.
Related MCP Servers
- AlicenseAqualityAmaintenanceEnables AI models to access and search offline Wikipedia and other knowledge bases stored in ZIM format files. Provides intelligent content retrieval, structured browsing, advanced search capabilities, and metadata extraction for comprehensive offline knowledge access.8285 PyPI143MIT
- AlicenseNot gradedqualityDmaintenanceEnables interaction with .zim archives by providing tools for article search, content retrieval, and metadata discovery. It features a TF-IDF based RAG engine for semantic retrieval over extracted article chunks from compressed ZIM files.GPL 3.0
- AlicenseNot gradedqualityDmaintenanceEnables large language models to directly access and search content in ZIM files, allowing offline question answering and information retrieval from resources like Wikipedia.22MIT
- AlicenseNot gradedqualityDmaintenanceProvides offline search and retrieval of Wikipedia articles using Kiwix .zim files, enabling LLMs to access full Wikipedia content without internet.2MIT









