I built a self-hosted AI knowledge base on top of offline Wikipedia (and it speaks MCP)

说明与用途

基于离线ZIM文件的自托管AI知识库，集成Claude MCP，并支持自动更新。

问题

The existing kiwix-serve tool only allows reading offline ZIM files; the author needed to query the content, keep it updated automatically, and use it as a source for local LLMs.

结果

Zim Pro transforms a read‑only offline ZIM collection into an interactive, AI‑powered knowledge base that can be queried, updated, and used by local LLMs without internet access.

工作流

• 1. Install Zim Pro (Docker, binary or pip).
• 2. Mount ZIM files into the /data volume or add them via the catalog.
• 3. Start the server and access the UI at http://localhost:8083.
• 4. Use the built‑in AI "Ask AI" feature to query the library; the system performs plan, deep search, light search, and rank steps before feeding cleaned Markdown to the model.
• 5. Integrate with Claude via MCP using claude mcp add --transport http zim-pro http://localhost:8083/mcp to expose tools like list_books, search, get_article, etc.
• 6. Configure per‑app API keys for secure access if needed.
• 7. Rely on auto‑updates to keep ZIMs current with the Kiwix catalog.

证据

On my home server I have a folder containing ZIM files: these are offline versions of Wikipedia, Stack Exchange, DevDocs and iFixit, all drawn from the Kiwix catalog.

来源

The problem was that kiwix-serve only allows you to read it. What I wanted to do was: ask it questions and get answers I could verify, keep the books updated without checking for new releases by hand, have my other AI tools use it as a source.

来源

Zim Pro, a drop‑in substitute for kiwix-serve which does all three things.

来源

The model receives articles in the form of cleaned Markdown, not HTML. Any navigation links, reference lists, "See also" sections, citation markers and licence footers are removed, and each passage is then given its section heading.

来源

Using it from Claude via MCP: claude mcp add --transport http zim-pro http://localhost:8083/mcp

来源