ZIM MCP Server

What is ZIM MCP Server?

ZIM MCP Server enables large language models to directly access and search content within ZIM files, allowing local AI models to perform question answering and information retrieval from offline knowledge resources like Wikipedia without internet access.

How to use ZIM MCP Server?

Install the uv package manager, clone the repository and run uv sync, then download ZIM files from the Kiwix Library and place them in a directory. Configure the server by adding the correct path to your claude_desktop_config.json with the uv --directory ... run server.py ... command and the ZIM directory as an argument.

Key features of ZIM MCP Server

• List all ZIM files in allowed directories
• Search within ZIM file content with a query term
• Get detailed content of a specific ZIM entry
• Supports pagination with limit and offset parameters
• Works fully offline with local AI models
• Integrates with Claude Desktop via MCP protocol

Use cases of ZIM MCP Server

• Enabling offline question answering from Wikipedia and other reference content
• Providing local AI models with access to large offline knowledge bases
• Accessing educational resources in environments without internet connectivity
• Searching and retrieving specific articles from ZIM archives programmatically

FAQ from ZIM MCP Server

What is a ZIM file?

ZIM (Zeno IMproved) is a file format developed by Kiwix for offline storage and access of Wikipedia and other large reference content, supporting high compression and fast searching.

What tools does ZIM MCP Server provide?

It provides three tools: list_zim_files (lists ZIM files in allowed directories), search_zim_file (searches within ZIM file content), and get_zim_entry (gets detailed content of a specific entry).

How do I configure Claude Desktop to use this server?

Add an entry to claude_desktop_config.json with the command uv, arguments containing --directory followed by the repository path, run server.py, and the directory containing your ZIM files.