World's first cross-source credibility-scored UAP database

Every Government
UAP Record.
One Graph.

361,491 sighting records spanning 38 independent sources — from US Air Force Project Blue Book to the Spanish CEFAE to GEIPAN France to the Hatch UDB and Lissoni Archives — unified, scored, and cross-linked for the first time in history.

Open the Explorer View All Sources
361,491
Total Records
26
Independent Sources
17,126
Cross-Source Links
66
Countries
1860
Earliest Record
9.0
Max Credibility Score
What is a Knowledge Graph

Data Connected,
Not Just Collected

A knowledge graph is a network where entities are nodes and relationships are edges. Most UAP databases treat records as isolated rows. This platform treats them as a living graph — where a Blue Book Air Force report from 1962 can be linked to a NUFORC civilian call about the same event 40 years later.

⬡
Nodes — Individual Sighting Records
Each of the 361,491 records is a node in the graph, carrying structured attributes: date, location, shape, credibility score, source tier, and evidence types.
⟷
Edges — Cross-Source Corroborations
The Haversine geospatial algorithm identifies when two independent sources document the same event within 50km and 30 days. Each confirmed match is an edge in the graph — 17,126 verified edges at the standard threshold.
◈
Credibility — Edge-Weighted Scoring
A civilian report that is corroborated by an Air Force investigation scores dramatically higher than either source alone. The graph structure elevates quality through connection — not through assertion.
Why This Is Different

Six Firsts in UAP Research

01
First Unified Cross-Government Database
For the first time, records from US (Blue Book, CIA, AARO, DIA, NSA), UK (MoD), France (GEIPAN), Spain (CEFAE), and Brazil (SIAN) exist in a single queryable schema. No aggregator has done this before.
02
Algorithmic Cross-Source Corroboration
17,126 verified events where two or more independent sources documented the same incident — computed for the first time using Haversine geospatial matching at 50km / 30-day thresholds against 800,000+ record pairs.
03
Structured Credibility Scoring Algorithm
Every record receives a 1.0–9.0 credibility score built from five independent dimensions: source tier, physical evidence, duration, geospatial precision, and cross-source corroboration. Reproducible, transparent, versioned.
04
AI-Assisted PDF Extraction at Scale
Over 800 government FOIA documents — previously unsearchable scanned PDFs — were processed using Claude Haiku via structured extraction prompts. 525 CIA records and 13 AARO analytical documents extracted this way.
05
Public API with Canned Query Endpoints
A live public JSON API serves 10 pre-defined research queries — high credibility cases, government sources, corroborated events, nuclear facility sightings — with no authentication required for read access.
06
MCP-Compatible AI Query Layer
An MCP server at uap-mcp-server.onrender.com allows any Claude-compatible AI to query the full database in natural language — the first UAP database to be AI-queryable via standardised protocol.
How It Works

The Pipeline

01 —
Source Acquisition
Each source is acquired from its primary origin: NARA FOIA releases, government open data portals, academic archives, or structured web scraping with rate limiting and robots.txt compliance. No third-party aggregators.
02 —
Structured Extraction
CSV sources are parsed with column normalisation and shape vocabulary mapping. PDF sources (800+ documents) are extracted page-by-page using pdfplumber, then processed by Claude Haiku with structured JSON prompts tuned to each document format.
03 —
Geocoding & Normalisation
Locations are resolved to latitude/longitude via OpenStreetMap Nominatim with fallback string matching. Dates are normalised to ISO 8601 with confidence levels (exact / date / month / year). Shapes are mapped to a canonical 15-category vocabulary.
04 —
Credibility Scoring
Each record receives a score from 1.0–9.0 across five dimensions: source tier (government vs. civilian), physical evidence types, observation duration, geospatial precision, and cross-source corroboration bonus. Algorithm is versioned and logged per ingestion run.
05 —
Geospatial Cross-Linking
The Haversine linker runs against all geolocated events with confirmed dates. For every source pair, it identifies events within configurable radius (default 25km) and time window (default ±7 days). Confidence is scored 60% geographic / 40% temporal. Pairs above 0.7 confidence are stored as same_event links.
06 —
Auditability & Versioning
Every ingestion run is logged with script version, source file hash, record count, and error summary. The database schema and scoring algorithm are versioned. The full pipeline is reproducible from raw source files.
Credibility Score Breakdown
Source Authority
1.0 – 6.0
Physical Evidence
+0.5 – 1.5
Duration
+0.5 – 1.0
Geospatial Precision
+0.5
Cross-Source Corroboration
Up to +4.0
Score Thresholds
● 9.0 Maximum
Gov + radar + corroborated
● 8.0–8.9 Very High
Gov + strong evidence
● 7.0–7.9 High
Gov or corroborated civilian
● 5.0–6.9 Medium
Expert civilian, good evidence
The Core Finding
17,126
verified cross-source corroborations
Cases where two or more completely independent databases documented the same incident at the same location within 30 days of each other — computed for the first time across 800,000+ record pairs. A civilian phone-in report and an Air Force investigation, unknowingly recording the same event decades apart.
Explore corroborated cases →
Top corroborated source pairs
NARA-FAA
↔
NUFORC
9,318 links
NUFORC
↔
NUFORC+
9,318 links
AARO
↔
CIA
3 links
Blue Book
↔
CIA
3 links
Blue Book
↔
NUFORC
8 links
GEIPAN
↔
NUFORC
2 links
Hatch-UDB
↔
NUFORC
653 links
Blue Book
↔
Hatch-UDB
190 links

Start Exploring the Graph

The interactive map loads 5,500 high-credibility events on first view. Filter by source, score range, date, shape, and geographic area. Click any marker to see full case details and cross-source links.

Open the Explorer Data Sources Request API Access
UAP Intelligence
361,491 records · 38 sources · MCP-powered
UAP Intel
Intelligence interface online. I have direct access to the UAP Knowledge Graph — 361,491 records across 34 government and civilian sources. Ask me anything: incidents, geographic patterns, shape analysis, credibility scoring, or cross-source corroboration.