OpenZIM

v1.0.0Knowledge & Memorystable

Enables 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 offlin

kiwixmcpmcp-serveropenzimzim
Share:
59
Stars
0
Downloads
0
Weekly
0/5

What is OpenZIM?

OpenZIM is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables 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...

Enables 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 offlin

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables AI models to access and search offline Wikipedia and

Use Cases

Access offline Wikipedia and knowledge bases stored in ZIM format.
Search and retrieve content without internet connection.
Extract structured data from offline resources.
cameronrye

Maintainer

LicenseMIT License
Languagepython
Versionv1.0.0
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx openzim-mcp-server

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use OpenZIM

The OpenZIM MCP server gives AI models access to offline knowledge bases stored in ZIM format files — the standard used by Kiwix to distribute Wikipedia, Project Gutenberg, and thousands of other reference collections. It provides both a simple natural-language query mode and eight specialized tools for searching, browsing articles, extracting sections, retrieving links, and inspecting metadata — all without requiring an internet connection. Researchers, students in low-connectivity environments, and anyone working with curated offline knowledge bases use it to query large reference collections through an AI assistant.

Prerequisites

  • Python 3.9+ with pip or uv installed
  • At least one ZIM file downloaded from https://library.kiwix.org (e.g., Wikipedia, Project Gutenberg)
  • An MCP client such as Claude Desktop that supports stdio MCP servers
  • Sufficient disk space for ZIM files (Wikipedia English is ~90 GB; smaller language editions are available)
1

Install the openzim-mcp package

Install using uv (recommended for isolated environments) or pip.

# Recommended:
uv tool install openzim-mcp

# Alternative with pip:
pip install openzim-mcp

# Verify the install:
openzim-mcp --help
2

Download a ZIM file

Go to https://library.kiwix.org and download a ZIM file for the knowledge base you want to query. Note the local file path — you will pass it to the server as an argument.

# Example: download a small Wikipedia edition
wget https://download.kiwix.org/zim/wikipedia/wikipedia_en_simple_maxi_2024-01.zim -P ~/zim-files/
3

Test the server from the command line

Run the MCP server directly to confirm it can read the ZIM file before connecting your client.

openzim-mcp ~/zim-files/wikipedia_en_simple_maxi_2024-01.zim
4

Configure your MCP client

Add the OpenZIM server to your claude_desktop_config.json with the path to your ZIM file directory or specific file.

{
  "mcpServers": {
    "openzim": {
      "command": "openzim-mcp",
      "args": ["/Users/you/zim-files"]
    }
  }
}
5

Enable advanced mode for granular tool access (optional)

By default the server exposes a single zim_query tool. To access all eight specialized tools (zim_search, zim_get, zim_get_section, zim_browse, zim_metadata, zim_links, zim_health), start the server with the advanced flag.

openzim-mcp --advanced /Users/you/zim-files

OpenZIM Examples

Client configuration

Claude Desktop config for the OpenZIM MCP server pointing at a directory of ZIM files.

{
  "mcpServers": {
    "openzim": {
      "command": "openzim-mcp",
      "args": ["/Users/you/zim-files"]
    }
  }
}

Prompts to try

Example prompts to use with the OpenZIM MCP server once connected.

- "Summarize the Wikipedia article on photosynthesis from my offline library"
- "Search for articles about quantum computing in the ZIM file"
- "Get the introduction section of the Wikipedia article on Marie Curie"
- "List all the links in the article on the French Revolution"
- "What ZIM files do I have available and what are their metadata details?"

Troubleshooting OpenZIM

Server exits immediately with 'no ZIM files found'

Pass the path to either a specific .zim file or a directory containing .zim files as the first positional argument. Ensure the path exists and the files have a .zim extension. Run openzim-mcp --help to confirm the expected argument format.

Slow search or article retrieval on large ZIM files

Large ZIM files (e.g., full English Wikipedia) require significant RAM for index operations. Ensure you have at least 4 GB of free memory. Consider using a smaller language edition or a topical ZIM (e.g., Wikipedia medicine) for faster queries.

HTTP transport mode authentication fails

When using HTTP transport (OPENZIM_MCP_TRANSPORT=http), set OPENZIM_MCP_AUTH_TOKEN to the same bearer token you pass in client requests. The default port is 8000; verify no other service occupies it. For local stdio usage no authentication is needed.

Frequently Asked Questions about OpenZIM

What is OpenZIM?

OpenZIM is a Model Context Protocol (MCP) server that enables 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 offlin It connects AI assistants to external tools and data sources through a standardized interface.

How do I install OpenZIM?

Follow the installation instructions on the OpenZIM GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with OpenZIM?

OpenZIM works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is OpenZIM free to use?

Yes, OpenZIM is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "openzim-mcp-server": { "command": "npx", "args": ["-y", "openzim-mcp-server"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use OpenZIM?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides