Data API Documentation

Every endpoint below is a static file generated at build time and served from a CDN. There is no authentication, no API key, and no rate limit beyond the CDN’s own protections. U.S. government source documents are generally public domain (17 U.S.C. §105); this catalog’s metadata is provided as-is. Cataloged by the independent UAP Records Archive — not an official or government source.

Endpoints

URL	Format	Contents	Count / source of truth
/uap/data/uap-records.json	application/json	Full record catalog: dataset name, license, disclaimer, count, records[] (fields below).	All published records — the payload’s own count field is the source of truth (80 at time of writing).
/uap/data/uap-records.csv	text/csv	Same catalog flattened to one row per record; header row id,title,agencyId,recordType,decade,status,topics,officialLandingUrl,sourceUrl,archiveUrls,disclaimer. Array fields are ;-joined.	All published records — one data row per record.
/uap/data/uap-records.bib	text/plain (BibTeX)	Bulk citation export: one BibTeX entry per record, linking the archive page and the source landing page.	All published records — one entry per record.
/uap/data/uap-records.ris	text/plain (RIS)	Bulk citation export in RIS, importable by Zotero / EndNote.	All published records — one entry per record.
/uap/data/uap-media.json	application/json	Inventory of media hosted on this site (videos, documents, images, audio): dataset name, license, disclaimer, count, media[] with type, record, url, label (and poster for video items).	All hosted media items — see the payload’s count field.
/uap/data/uap-media.csv	text/csv	Same inventory as CSV; header row type,record,url,label,disclaimer.	All hosted media items — one data row per item.
/uap/data/checksums.txt	text/plain (sha256sum)	SHA-256 checksums for every dataset artifact and feed above, regenerated on each build; sha256sum -c compatible. Also available as JSON at /uap/data/checksums.json ({ generated, algorithm, files }).	One line / entry per dataset artifact (12 files).
/feed.xml	application/rss+xml (RSS 2.0)	Site-wide feed: one item per published record (title, link, description).	All published records.
/feed.json	application/feed+json (JSON Feed 1.1)	Same feed as JSON Feed 1.1; items carry external_url pointing at the source landing page where available.	All published records.
`/feed/{agencyId}.xml`	application/rss+xml (RSS 2.0)	Per-agency feed, filtered to one agency’s records. agencyId is one of: aaro, nara, nasa, odni (e.g. /feed/nara.xml).	Published records of that agency only.
`/uap/records/{id}.json`	application/json	Single record by id: record (fields below), canonical page URL, license, disclaimer. See example payload below.	One endpoint per published record; ids match the id field in uap-records.json.
/search-index.json	application/json (array)	Build-time search index over catalog pages (records, agencies, timeline; en + zh): id, collection, title, slug, url, locale, description, agency, type, decade, status, topics.	One entry per searchable catalog page — the array length is the count.
/data.json	application/json (DCAT)	Project Open Data / DCAT catalog metadata describing the datasets above — discovery metadata for data portals, not record data itself.	One catalog document.
/pagefind/	pagefind index (fragmented, on-demand)	Full-text inverted index over reviewed extracts and machine-extracted document texts, chunked per source page. Powers the site search; consumable from any page via the pagefind JS API (/pagefind/pagefind.js). Replaced the former monolithic /doc-search-index*.json endpoints (retired 2026-06-12).	Fragmented — the browser loads only the fragments a query needs.
/uap/data/cases.json	application/json	Case layer: verbatim official case designations (PURSUE file titles, quoted AARO record titles) with their grouped hosted assets (documents, videos, images, audio, records) and verbatim incident dates/locations from the official catalog. Groups by official designation only; asserts no conclusions.	The top-level "count" field (one entry per official case designation).

Per-record endpoint

Each published record is also available as a standalone JSON file at /uap/records/{id}.json. Example — /uap/records/aaro-mission.json (truncated):

{
  "record": {
    "id": "aaro-mission",
    "title": "AARO Mission & Agency Page",
    "agencyId": "aaro",
    "recordType": "agency_landing",
    "decade": "2020s",
    "status": "official_link_only",
    "topics": [
      "agency-position-page",
      "aaro-case-resolution"
    ],
    "officialLandingUrl": "https://www.aaro.mil/",
    "sourceUrl": "https://www.aaro.mil/",
    "archiveUrls": ["…"]
  },
  "canonical": "https://uap-archive.org/uap/records/aaro-mission/",
  "license": "…",
  "disclaimer": "…"
}

Record field reference

The record object is identical in uap-records.json, uap-records.csv, and the per-record endpoint.

id: Unique slug identifier for the record (also the {id} of the per-record endpoint).
title: Short descriptive title of the record.
agencyId: Identifier of the originating agency. One of: nara, nasa, odni, aaro.
recordType: Category of the record. One of: agency_landing, collection_landing, report_landing, imagery_landing.
decade: Decade of the source document. One of: 1940s, 1970s, 1980s, 2000s, 2010s, 2020s.
status: Verification status in this archive release. One of: official_link_only, needs_review.
topics: Topic tags from the archive’s controlled vocabulary (string array in JSON; ;-joined in CSV). May be empty.
officialLandingUrl: URL of the official landing page for the source document.
sourceUrl: Direct URL to the source document (PDF, video, or web page).
archiveUrls: Third-party archive copies, e.g. Wayback Machine (string array in JSON; ;-joined in CSV). May be empty.

Usage examples

curl

# Full record catalog (JSON; payload carries license, disclaimer, count, records[])
curl -s https://uap-archive.org/uap/data/uap-records.json

# One record by id
curl -s https://uap-archive.org/uap/records/aaro-mission.json

# RSS feed for one agency (aaro | nara | nasa | odni)
curl -s https://uap-archive.org/feed/nasa.xml

JavaScript (fetch)

const res = await fetch('https://uap-archive.org/uap/data/uap-records.json');
const { count, records } = await res.json();
const reports = records.filter((r) => r.recordType === 'report_landing');
console.log(count, 'records total;', reports.length, 'report landing pages');

License & attribution

U.S. government source documents are generally public domain (17 U.S.C. §105); this catalog’s metadata is provided as-is.

Cataloged by the independent UAP Records Archive — not an official or government source.