mcpField
U
UAPDrop Research Archive

UAPDrop Research Archive

Remote
uapdrop.comSearch0Updated 11w ago

About UAPDrop Research Archive

UAPDrop Research Archive is a hosted MCP server for researchers and MCP-capable assistants exploring UAP/UFO records. It combines searchable government documents, geocoded sightings, videos, news, and cited entity profiles, with results linked to source records.

Key features

  1. 01Search declassified documents
  2. 02Retrieve full documents and sightings
  3. 03Filter geocoded sightings by source, date, and area
  4. 04Search analysed commentary videos
  5. 05Follow links between documents and videos
  6. 06Semantic search and cited entity profiles

Use cases

  1. 01Check original records behind a historical sighting
  2. 02Find sightings within a date range and area
  3. 03Explore material related to a commentary video
  4. 04Review citations behind reported entity claims

MCP support

The hosted server exposes read-only search and retrieval tools for documents, sightings, videos, news, and entity profiles to MCP clients. It uses streamable HTTP and bearer API-key authentication.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.uapdrop.com/the-truth-is-in-here
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://mcp.uapdrop.com/the-truth-is-in-hereAuth required

Tools

Details

From the project README

UAPDrop Research Archive — MCP Server

A remote Model Context Protocol server that gives Claude, ChatGPT, Cursor and any other MCP-capable client grounded, citable access to the UAPDrop research corpus: declassified UAP/UFO government documents, a geocoded sighting archive built from national archives, news coverage, and cited entity profiles.

Every result links back to the record it came from, so the assistant's claims can be checked against a primary source rather than taken on trust.

This repository holds the server's public metadata (server.json) and documentation. The server itself is hosted — there is nothing to install or run locally.

What's in the corpus

Declassified documents13,929 — Project Blue Book, the FBI Vault, and national-archive releases, with extracted PDF text
Geocoded sightings28,314 records from four vetted government sources
Sighting sourcesUS NARA Project Blue Book, UK National Archives, France's CNES/GEIPAN, and the Hatch catalogue
Commentary videos~10,000 analysed, each with a generated summary, analysis and talking points
Entity profilesReported non-human entities, every claim backed by a citation into the archive

The underlying sighting dataset is free to download in bulk and always will be. What this server provides is the retrieval layer over it — keyword search, semantic search, ranking, and cross-linking across the whole corpus.

Tools

All tools are read-only. There are no ingest, generation, or administrative tools, by design.

ToolWhat it doesPlan
search_documentsFull-text search across the declassified archive — title, summary, incident text, tags and extracted PDF text. Filter by tag, releasing agency, or file type.Free
get_documentOne archive document in full, including its attachment gallery.Free
search_sightingsThe geocoded sighting archive, filterable by source, date range and bounding box.Free
get_sightingOne sighting record with its full media set.Free
list_sighting_sourcesThe source keys available to search_sightings.Free
search_videosKeyword search over analysed commentary videos.Free
get_videoOne commentary video with our written summary, analysis and talking points.Free
related_documentsArchive documents related to a video, diversified across release programs.Free
related_commentaryCommentary videos related to an archive document.Free
search_semanticEmbedding-based search across documents, news and sightings — finds conceptually related material that keyword search misses.Researcher
list_news / get_newsUAPDrop news coverage, including related-commentary links.Researcher
list_beings / get_beingReported entity profiles with their full cited-references ledger.Researcher

Raw video transcripts are never returned, and neither is YouTube's own description text.

Setup

Create a key on your profile page — sign-in is free and so is the starter tier — then register the server with your client.

Claude Code
claude mcp add uapdrop --transport http \
  https://mcp.uapdrop.com/the-truth-is-in-here \
  --header "Authorization: Bearer uapd_live_YOUR_KEY_HERE"
Claude Desktop, Cursor, VS Code, and other file-configured clients
{
  "mcpServers": {
    "uapdrop": {
      "type": "http",
      "url": "https://mcp.uapdrop.com/the-truth-is-in-here",
      "headers": { "Authorization": "Bearer uapd_live_YOUR_KEY_HERE" }
    }
  }
}
Verifying the connection

The server publishes OAuth 2.0 Protected Resource Metadata, so an unauthenticated request returns a spec-conformant challenge rather than a bare failure:

curl -s https://mcp.uapdrop.com/.well-known/oauth-protected-resource

Plans

FreeResearcher
Price$0$50/month, cancel any time
Calls per day505,000
Document and sighting archives✅✅
Cross-links and video search✅✅
Semantic search across the corpus—✅
News archive and entity profiles—✅

Quotas are per key and reset at 00:00 UTC. Responses carry X-RateLimit-Remaining; exceeding the quota returns 429 with a Retry-After header.

Example prompts

Once connected, these are the kinds of questions the server is built to answer:

  • "What did Project Blue Book conclude about the 1952 Washington D.C. sightings, and what do the original documents actually say?"
  • "Find sightings within 100 km of Rendlesham Forest between 1980 and 1981, and show me the source records."
  • "Search the archive for anything conceptually similar to 'metallic disc observed by military radar operators' — not just those keywords."
  • "What primary sources back the claims in the entity profile for the Hopkinsville goblins?"

Data provenance

Records come from public government archives — the US National Archives' Project Blue Book, the UK National Archives, France's CNES/GEIPAN, and other national disclosure releases. Every result carries the URL of the originating record so it can be verified against the source.

Links

License

The contents of this repository (documentation and server metadata) are released under the MIT License. The archive data itself is drawn from public government sources; see the dataset page for its licensing.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.