I built a self-hosted AI knowledge base on top of offline Wikipedia (and it speaks MCP)
说明与用途
Zim Pro 替代 kiwix-serve,提供搜索、AI 回答(带引用)、离线运行、自动更新和 MCP 集成,使本地 LLM 能在无网络环境下使用。
问题
kiwix-serve only allows reading offline ZIM files; the author wanted to ask questions, get verifiable answers, and keep books updated automatically while using local LLMs.
结果
Zim Pro enables offline AI knowledge base with search, AI answering, MCP integration, auto‑updates, and custom ZIM creation, all without internet exposure.
工作流
- 1. Install Zim Pro via Docker or pip.
- 2. Mount ZIM files into /data and state into /state.
- 3. Start the container and access http://localhost:8083.
- 4. Use built‑in catalog or drop ZIM files into zims/ folder.
- 5. Configure MCP transport with "claude mcp add --transport http zim-pro http://localhost:8083/mcp".
- 6. Use MCP tools (list_books, search, suggest, get_article, get_article_outline, list_updates) from Claude or other LLMs.
- 7. Zim Pro automatically checks for updates against Kiwix catalog and downloads new versions with multi‑connection, checksummed, resumable downloads.
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.
The problem was that kiwix-serve only allows you to read it.
Zim Pro ... does all three things.
Ask AI: this will provide answers that are based solely on your books, including numbered citations referring back to the original articles, together with the images and videos from those pages.
Auto‑updates: for each book it is compared with the Kiwix catalog; the downloads make use of multiple connections, are checksummed in parts, and can be resumed if there is a crash.
Using it from Claude via MCP: claude mcp add --transport http zim-pro http://localhost:8083/mcp
- 记录日期: 2026-10-16T08:55:59.601Z