互联网档案搜索下载工具

mcp-internet-archive

MCP server for searching and downloading from the Internet Archive via the ia CLI.

Exposes four read-only tools that let Claude search items, read metadata, list files, and download content from archive.org.

Prerequisites

• Node.js 18+
• Internet Archive CLI (ia)

Installing the ia CLI

pip install internetarchive
ia configure   # optional — needed for some restricted items
ia --version   # verify installation

Installation

Install from GitHub (recommended)

No need to clone — install directly and use with npx:

npm install -g github:lawriec/mcp-internet-archive

Or without installing globally, just reference it in your Claude config (see below).

Install from source

git clone https://github.com/lawriec/mcp-internet-archive.git
cd mcp-internet-archive
npm install
npm run build

Claude Desktop Configuration

Add to your Claude Desktop config (%APPDATA%\Claude\claude_desktop_config.json on Windows, ~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

Using npx (no clone needed):

{
  "mcpServers": {
    "internet-archive": {
      "command": "npx",
      "args": ["-y", "github:lawriec/mcp-internet-archive"]
    }
  }
}

Using a local clone:

{
  "mcpServers": {
    "internet-archive": {
      "command": "node",
      "args": ["C:/Users/youruser/dev/mcp/mcp-internet-archive/build/index.js"]
    }
  }
}

No environment variables needed — read-only operations don't require authentication.

Tools

ia_search

Search archive.org items using Lucene query syntax.

Example queries:

• collection:prelinger subject:"san francisco"
• creator:"Mark Twain" mediatype:texts
• title:"Apollo 11"

ia_metadata

Get full metadata for a specific item.

ia_download

Download files from an item to a local directory.

ia_list

List all files in an item with metadata (name, size, format, checksums).

Troubleshooting

"The Internet Archive CLI (ia) is not installed or not on PATH"

• Install with pip install internetarchive
• Ensure the ia command is on your system PATH
• Verify with ia --version

Timeouts on large downloads

• Downloads have a 10-minute timeout. For very large items, use glob or format to download specific files.

Empty search results

• Check your query syntax at archive.org/advancedsearch.php
• Try broader queries or different field names

License

MIT