loc-mcp-server by wmo4buva - MCP Server

Library of Congress MCP Server

A Model Context Protocol (MCP) server providing access to the U.S. Library of Congress digital collections API. This server enables Claude and other MCP clients to search, browse, and retrieve information from the Library of Congress's vast digital archives.

Features

🔍 Universal Search: Search across all Library of Congress digital collections
📚 Format-Specific Search: Search within specific material types (books, audio, photos, maps, etc.)
🏛️ Collection Browsing: Browse and explore digital collections
📄 Item Details: Get detailed information about specific items
🔗 Resource Access: Access individual pages and segments of items
📊 Advanced Filtering: Filter results by location, subject, contributor, and more

Available Tools

`loc_search`

Search across all Library of Congress digital collections.

Parameters:

query (optional): Search query string
facets (optional): Array of facet filters (e.g., ["location:ohio", "subject:baseball"])
count (optional): Results per page (1-100, default 25)
page (optional): Page number for pagination
sort (optional): Sort by date, title, or shelf ID (with _asc/_desc)
attributes (optional): Specific attributes to return

`loc_search_format`

Search within a specific material format.

Parameters:

format (required): One of: audio, books, film-and-videos, journals, maps, manuscripts, music, newspapers, notated-music, photos, prints, software, web-pages
query (optional): Search query string
Other parameters same as loc_search

`loc_get_collections`

Get a list of all available digital collections.

Parameters:

count (optional): Collections per page
page (optional): Page number

`loc_get_collection`

Get information about a specific collection and browse its contents.

Parameters:

collectionName (required): Collection identifier
query (optional): Search within the collection
count (optional): Items per page
page (optional): Page number

`loc_get_item`

Get detailed information about a specific item.

Parameters:

itemId (required): Item identifier
attributes (optional): Specific attributes to return

`loc_get_resource`

Get information about a specific resource (page/segment of an item).

Parameters:

resourceId (required): Resource identifier
attributes (optional): Specific attributes to return

Installation

npm install
npm run build

Integration with AI Desktop Applications

This MCP server works with any application that supports the Model Context Protocol (MCP).

Claude Desktop

Add to your Claude Desktop configuration file (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "loc": {
      "command": "node",
      "args": ["/path/to/loc-mcp-server/dist/index.js"],
      "env": {
        "NODE_ENV": "production"
      }
    }
  }
}

Or use npx (after publishing to npm):

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

OpenAI ChatGPT Desktop

Add to your ChatGPT Desktop configuration file:

macOS: ~/Library/Application Support/ChatGPT/config.json Windows: %APPDATA%/ChatGPT/config.json Linux: ~/.config/ChatGPT/config.json

{
  "mcpServers": {
    "loc": {
      "command": "node",
      "args": ["/absolute/path/to/loc-mcp-server/dist/index.js"],
      "env": {
        "NODE_ENV": "production"
      }
    }
  }
}

Note: Replace /absolute/path/to/loc-mcp-server with the actual path where you cloned this repository.

Other MCP-Compatible Applications

This server follows the standard MCP protocol and should work with any MCP-compatible application. Refer to your application's documentation for MCP server configuration instructions.

Example Queries

Search for Abraham Lincoln materials:

Use the loc_search tool to find materials related to Abraham Lincoln

Find historical photographs:

Use the loc_search_format tool to search for "civil war" in the "photos" format

Browse a specific collection:

Use the loc_get_collection tool to explore the "american-folklife-center" collection

Get item details:

Use the loc_get_item tool to get detailed information about a specific item ID

API Information

This server uses the Library of Congress JSON/YAML API:

Base URL: https://www.loc.gov/
Documentation: https://www.loc.gov/apis/json-and-yaml/requests/
No authentication required

Rate Limiting

Important: The Library of Congress API has strict rate limits:

20 requests per minute maximum
1 hour blocking if rate limit is exceeded
429 status codes trigger automatic retry with exponential backoff

This MCP server implements:

✅ Automatic rate limiting (20 requests/minute)
✅ Request queuing and spacing
✅ 429 status code handling with retry logic
✅ Exponential backoff for failed requests
✅ Automatic window reset every minute

Available Material Formats

audio: Audio recordings and spoken word
books: Books and texts
film-and-videos: Motion pictures and videos
journals: Periodicals and serials
maps: Maps and cartographic materials
manuscripts: Handwritten and typescript materials
music: Musical recordings and scores
newspapers: Newspaper collections
notated-music: Sheet music and scores
photos: Photographs and images
prints: Prints and graphic arts
software: Digital software and programs
web-pages: Archived web content

Common Facet Filters

Use these in the facets parameter to filter results:

location:state_name (e.g., "location:virginia")
subject:topic (e.g., "subject:baseball")
contributor:name (e.g., "contributor:lincoln")
language:code (e.g., "language:english")
date:year (e.g., "date:1865")

Development

# Install dependencies
npm install

# Build
npm run build

# Run in development mode
npm run dev

# Start the server
npm start

License

MIT

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.