🗄️ NARA Catalog Extractor

Search the US National Archives — get download URLs. Give a keyword (or a specific record/ collection naId) and get every digital object's direct download URL plus metadata, as structured JSON. The only dedicated NARA-catalog extractor on the Store.

Headless · keyless · structured JSON out · built for AI agents.

⚡ Why it's different

The other government actors wrap the Federal Register or Data.gov — not the archival catalog itself. This one searches NARA and hands you the actual file URLs (PDF/TIFF/image/text), ready to download or chain into a PDF/OCR extractor.

📋 What you get (one row per digital object)

Field	Meaning
naId	The record's National Archives ID
title	Record title
recordUrl	Catalog page for the record
objectFilename	The file's name
objectUrl	Direct download URL for the file
fileExtension, fileType	.pdf/.tif/… and readable / media / record

📥 Input

Field	Type	Meaning
query	string	Keyword search (use this or naId), e.g. UAP
naId	string	A specific record/collection; collections page their descendants automatically
includeMedia	boolean	Include audio/video (default: readable only)
maxObjects	integer	Optional cap on rows
maxRecords	integer	Keyword search: how many records to pull (default 25)
fileTypes	array	Only return objects with these extensions, e.g. ["pdf","tif"]
webhookUrl	string	POST results here (Zapier/Make/n8n/custom)

Example input

{ "query": "UAP", "maxRecords": 5 }

📤 Output (one row per digital object)

{
  "naId": "499915731", "title": "UAP 10/26/2024",
  "recordUrl": "https://catalog.archives.gov/id/499915731",
  "objectFilename": "237UAP00386.pdf",
  "objectUrl": "https://catalog.archives.gov/.../237UAP00386.pdf",
  "fileExtension": ".pdf", "fileType": "readable", "capturedAt": "2026-..."
}

💡 Use cases

• Researchers / journalists / genealogists: pull an entire collection's files in one call.
• RAG / AI pipelines: get the file URLs, then chain to a PDF/OCR extractor to read them.
• FOIA / OSINT: turn a topic search into a downloadable document set.
• Archival backups: enumerate a collection's digital objects programmatically.

🤖 For AI agents (MCP)

Keyless, one call → the record list with direct file URLs, ready to chain. Example MCP config:

{
  "mcpServers": {
    "nara-catalog": {
      "url": "https://mcp.apify.com/?tools=actors,docs,centorprat/nara-catalog-extractor"
    }
  }
}

Claude Code: claude mcp add --transport http apify "https://mcp.apify.com/?tools=actors,docs,centorprat/nara-catalog-extractor" --header "Authorization: Bearer <YOUR_APIFY_TOKEN>"

No extra setup: this Actor is callable from Claude, Cursor, ChatGPT or any MCP client through the Apify MCP server, and agents can find it on their own with the server's search-actors tool.

🚀 Ready-made tasks (one click)

Pre-configured runs for the most common jobs. Open one, change the input, run:

• UAP & UFO Records from the US National Archives: Find declassified UAP and UFO records in the US National Archives catalog with direct download links for every digitized file.
• JFK Assassination Records: Download Links (NARA): Direct download links for JFK assassination records in the US National Archives catalog, with titles, dates and record hierarchy.
• US National Archives Photo Search (Download URLs): Search public-domain historical photos in the US National Archives and get direct JPG/TIFF download URLs with titles and dates.

❓ FAQ

• Do I need an API key? No — the NARA catalog API is public and keyless.
• Keyword or naId? query searches by topic; naId targets one record or a whole collection (its descendants are paged for you).
• What file types? PDF, TIFF/JPG/PNG images, text/JSON/CSV; audio/video only if includeMedia is on.
• A record with no files? Search hits with no digital objects return as a record-only row (so nothing is lost).
• How many results? Up to maxObjects; keyword mode pulls up to maxRecords records (20/page).
• Can I read the file contents here? This returns the URLs + metadata; chain to a PDF/OCR extractor for text.
• Is data stored? No — fetched, returned.
• Commercial use? NARA archival descriptions and US federal records are generally public domain.

📜 Data & attribution

Data from the US National Archives Catalog (catalog.archives.gov). Public catalog metadata and object pointers; nothing stored.

🔗 Related actors

Part of one suite of keyless, agent-ready composers by the same developer:

• Shopify Store Leads + App Audit (Reviews, Email, Pixels): Shopify stores by niche, verified and audited for missing apps (reviews, email/SMS, upsell, loyalty, chat, subscriptions), pixels and theme: scored leads with emails.
• Google Maps Leads + Website Audit for Agencies: local businesses from Google Maps audited for website, Google profile, reviews, booking and tracking gaps: scored, explained leads with phone and emails.
• FEMA Flood Zone & Property Risk API - Zillow/Redfin Enricher: FEMA flood zone + 18-hazard FEMA risk index + neighborhood demographics per US address, or for a whole Zillow/Redfin dataset.
• Domain Lookup API - DNS, MX, SPF, Tech & Contact Emails: DNS + live homepage → IPs, MX/SPF, tech, socials and public contacts per domain.
• Wikipedia & Wikidata API - Entity Facts, Summary & Pageviews: Wikipedia + Wikidata + pageviews → one structured brief per company, person or place.
• Country Economic Data API - World Bank GDP, FX & Holidays: World Bank + FX + public holidays → one market brief per country.
• Weather, Air Quality & Earthquake Risk API by Location: live weather + air quality + earthquakes → one clear/caution/hazard verdict per point, worldwide.
• Aircraft Tracker API (ADS-B) - Approach & Military Alerts: aircraft approaching a point with CPA/ETA + military/unidentified flags and alerts.