epheterson

epheterson / Zimi

Public

The offline internet — searchable, browsable, and self-updating. A modern server for ZIM files with cross-source search, built-in library management, and a JSON API.

api docker kiwix knowledge-base mcp
16
0
100% credibility
Found Feb 17, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

Zimi is a user-friendly offline search engine and reader for massive archives of Wikipedia, Stack Overflow, documentation, and other knowledge sources stored in ZIM files.

How It Works

1
🔍 Discover Zimi

You learn about Zimi, a simple way to download and search huge collections of articles from Wikipedia, coding sites, and more, all working offline on your computer.

2
🚀 Start Zimi

You get the free program and launch it on your machine, setting up a folder where your knowledge collections will live.

3
📚 Build Your Offline Library

Open the friendly web page in your browser, go to settings, and easily browse and download packs of articles like full Wikipedia or helpful guides.

4
🔎 Search Everything

Type in a question like 'how to fix a bike' and instantly see matching articles from all your collections, ranked by relevance.

5
📖 Dive into Articles

Click any result to read the full article in a clean reader view, with images and text, all speedy and private offline.

✅ Master Offline Knowledge

Enjoy lightning-fast searches and reading across millions of articles anytime, no internet needed, perfect for travel or spotty connections.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Zimi?

Zimi runs a local knowledge server for ZIM files—massive offline archives of Wikipedia, Stack Overflow, dev docs, and more from Kiwix. Fire up the Python Docker container, mount your ZIMs, and get instant JSON API access to search, read articles as plain text, autocomplete titles, or grab random entries. It includes a web UI for browsing, a library manager to download ZIMs from the Kiwix catalog, and CLI tools for quick queries like searching readthedocs content offline.

Why is it gaining traction?

Unlike online APIs with rate limits or costs, Zimi delivers sub-second title searches and full-text results across gigabytes of local data, scoped by collections like "Dev" for Stack Overflow plus devdocs. The MCP server hooks directly into Claude Code or other AI agents for tools like search or read, making offline knowledge programmable. Docker Compose setup and env vars for rate limiting or auto-updates keep it dead simple for self-hosting.

Who should use this?

AI devs building agents that search github code patterns or read emails in Gmail-like archives without net access. Backend engineers in secure envs needing to search github repositories, organizations, or readthedocs offline. Docs writers or learners querying limit laws in math ZIMs or planning reading lists locally.

Verdict

With 13 stars and 1.0% credibility score, Zimi is early-stage but battle-ready—solid docs, perf tests, and MIT license make it worth a spin for offline search needs. Try the Docker quickstart if you want private, AI-ready knowledge without vendor lock-in.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.