# Library of Congress MCP MCP > Library of Congress MCP connects your AI client to the digital archives of the United States' oldest federal cultural institution. Search millions of historical newspapers, rare maps, sound recordings, and manuscripts directly from your IDE or chat interface. Stop manually navigating catalog portals and start pulling primary source data on demand. ## Overview - **Category:** knowledge-management - **Price:** Free - **Tags:** digital-archives, historical-data, catalog-search, metadata, public-records, research-tools ## Description You are looking at a direct pipeline to the Library of Congress. This MCP gives your AI client the keys to millions of digitized historical records. Think about the last time you needed to track down a specific Civil War map or pull the full text of a 19th-century newspaper. You probably opened a browser, navigated a clunky web portal, clicked through three different filters, and copied text manually. That stops today. When you connect this through Vinkius, your agent handles the catalog navigation for you. You just ask for what you need. The agent queries the archives, filters by media type, and pulls the exact bibliographic metadata or high-resolution image coordinates you asked for. It handles the OCR text extraction and image metadata behind the scenes. You get raw, structured data back in your workspace. This is built for people who actually need to read the primary sources. Historians tracking down manuscript drafts, data scientists running text analysis on historical newspapers, or educators building lesson plans from original documents. Look, you do not need to learn the LOC API. You just talk to your agent. Keep in mind that while an API key is optional, adding one bumps up your rate limits significantly if you plan on running large batch queries. The bottom line is simple. You get instant access to the largest library in the world, delivered as structured JSON right where you work. ## Tools ### get_collection_items Lists every individual item contained within one specific digital collection. ### list_collections Returns a complete list of all available digital collections in the archive. ### search_format Restricts your search to a specific media type like maps, photos, or audio files. ### get_image_info Returns IIIF technical metadata and coordinates for high-resolution historical images. ### get_item Pulls detailed bibliographic metadata for a single historical record. ### get_resource Fetches a discrete digitized file, like a specific page of a historical newspaper. ### search Queries the entire Library of Congress website using a plain text keyword. ### get_text_service Extracts full-text OCR, word coordinates, and context snippets from a document. ## Prompt Examples **Prompt:** ``` Search the Library of Congress for Civil War maps of Virginia. ``` **Response:** ``` I've found several maps. Notable items include 'Map of the seat of war in Virginia' (1861) and 'Central Virginia: showing Lieut. Gen'l. U.S. Grant's campaign'. Would you like the bibliographic details for any of these? ``` **Prompt:** ``` List all items in the 'world-war-i-posters' collection. ``` **Response:** ``` Retrieving items from the World War I Posters collection... I've found items like 'I want you for U.S. Army' and 'Destroy this mad brute'. There are 25 items on the first page. Should I fetch more or get details for a specific poster? ``` **Prompt:** ``` Get the full text for the document segment 'service/gdc/gdcscd/00/01/02/03/0001.txt'. ``` **Response:** ``` Accessing text services... I have retrieved the full OCR transcript for that segment. It begins with: 'PROCEEDINGS OF THE CONVENTION...'. Would you like me to summarize the content or search for a specific term within it? ``` ## Capabilities ### Search the global catalog Query the entire LOC database using plain text keywords to find historical records. ### Filter by media format Restrict search results to specific types like maps, audio files, or photographs. ### Browse digital collections List and drill down into curated archives like the Abraham Lincoln Papers. ### Extract full-text OCR Pull raw text, word coordinates, and context snippets from digitized documents. ### Fetch high-res image metadata Get IIIF technical data and coordinates for large format historical images. ### Retrieve discrete files Download specific pages of a newspaper or individual manuscript folios directly. ## Use Cases ### Tracking down a specific map A historian needs Civil War maps of Virginia. They ask their agent to use search_format for maps with those keywords, then use get_item to pull the exact bibliographic details for the 1861 campaign map. ### Analyzing historical text A data scientist is studying 19th-century political conventions. They use get_text_service to pull the full OCR transcript and word coordinates from a specific document segment for their NLP pipeline. ### Building a visual curriculum An educator needs World War I propaganda posters. They use list_collections to find the poster archive, then get_collection_items to retrieve the first batch of items for their lesson plan. ### Getting high-res image data An archivist needs to render a massive historical photograph. They use get_image_info to fetch the IIIF technical metadata and exact image coordinates required by their viewing software. ## Benefits - You stop clicking through web portals. The search and search_format tools let your agent query millions of records using plain English, returning structured data directly to your chat. - You get exact file access. Instead of downloading entire collections, get_resource pulls the exact discrete file you need, like page four of an 1890 newspaper. - You skip manual text transcription. get_text_service extracts full OCR text and word coordinates from digitized manuscripts, handing you clean text for analysis. - You get proper image data. get_image_info fetches IIIF technical metadata, giving your agent the exact coordinates needed to render or crop high-res historical photos. - You navigate curated archives instantly. Use list_collections and get_collection_items to drill down into specific historical sets without guessing catalog numbers. ## How It Works The bottom line is you skip the web portal and pull primary source data straight into your editor. 1. Subscribe to the MCP and optionally add your Library of Congress API key for higher rate limits. 2. Connect your preferred AI client and ask it to find historical records, maps, or manuscripts. 3. Get structured bibliographic data, full OCR text, or image metadata returned directly to your workspace. ## Frequently Asked Questions **How do I search for only maps using the Library of Congress MCP?** Use the search_format tool instead of the standard search. It restricts the query to a specific media type, so you only get map results back. **Can the Library of Congress MCP extract text from old newspapers?** Yes. The get_text_service tool pulls full-text OCR, word coordinates, and context snippets directly from digitized historical documents. **Do I need an API key to use the Library of Congress MCP?** No, it works without one, but adding your own Library of Congress API key is highly recommended to get higher rate limits for heavy querying. **How do I get the technical metadata for a high-res image with the Library of Congress MCP?** Call the get_image_info tool with the image identifier. It returns the IIIF technical data and coordinates you need to render or crop the image. **Can the Library of Congress MCP list all available digital collections?** Yes, the list_collections tool returns a complete directory of all digital collections, which you can then drill into using get_collection_items. **How do I get a specific page of a newspaper using the Library of Congress MCP?** Use the get_resource tool with the specific resource ID for that page. It fetches the discrete digitized file without downloading the entire newspaper. **What are the rate limits when using the search tool in the Library of Congress MCP?** You get 100 requests per hour without an API key. Adding your personal Library of Congress API key during setup increases this limit. That lets you run broader historical queries without hitting throttling errors. **Can I use the get_item tool in the Library of Congress MCP with Cursor?** Yes. This MCP works with any compatible client, including Cursor, Claude, and VS Code. Just add the configuration to your client settings. Your agent pulls detailed bibliographic data directly into your workspace. **Can I search for items specifically in a format like 'maps' or 'audio'?** Yes! Use the `search_format` tool and specify the format parameter (e.g., 'maps', 'photos', 'audio') along with your query to get targeted results. **How do I get the full OCR text of a digitized document?** You can use the `get_text_service` tool. Provide the document segment path and set the `full_text` parameter to 1 to retrieve the complete OCR transcript. **Can I list items from a specific collection like the Abraham Lincoln papers?** Yes. Use the `get_collection_items` tool with the collection slug (e.g., 'abraham-lincoln-papers') to see all items contained within that specific archive.