ask your ai what's happening on the ground
Connect your AI assistant to Crowd Threat's verified, human-reported threat data. Ask questions in plain English and get answers that show the verification status and confidence of every incident.
Two incidents within 5km of your office:
The cordon report is still unverified and comes from a newer reporter, so treat it as an early signal.
every report on the map, one question away
The same verified incidents our reporters put on the Crowd Threat map flow through the MCP straight into your AI assistant. Ask about a region and it pulls the live reports, with status and confidence attached.
38 incidents, mostly security and conflict reports and military movement, plus maritime activity in the shipping lanes.
a direct line from your ai to our data
MCP (Model Context Protocol) is an open standard that lets AI assistants connect to outside data sources. Instead of copying and pasting reports into a chat, your AI looks up Crowd Threat data itself whenever your question needs it.
Your AI tool
You ask a question in plain English, in the assistant you already use.
Crowd Threat MCP
Turns your question into a search of our threat feed by place, time and category.
Verified threat data
Human-reported incidents from our global reporter network, with status and confidence.
ground truth, wherever you work
General AI tools answer from what they were trained on or what they can scrape. The Crowd Threat MCP gives yours live, verified reporting from people on the ground.
Verified, not scraped
Answers come from incidents reported by people on the ground, each carrying its live verification status and a confidence level.
In the tools you already use
No new dashboard to learn and no exports. Your AI assistant pulls Crowd Threat data straight into the conversation.
Ask in plain English
Filter by place, time and category just by asking. No API calls, no query builders, no code.
questions your ai can now answer
A few examples of what security, travel and risk teams can ask once they're connected.
"Any verified incidents within 5km of our Lagos office in the last 48 hours?"
"Summarise civil unrest reporting in Kenya over the last 30 days."
"What's been reported near these three hotels since Monday?"
"Draft this morning's security brief for our sites in Mexico City."
"Which of these reports are still unverified, and how confident should I be in them?"
"Compare reported protest activity in Paris and Berlin this week."
connected in minutes
If your AI tool supports MCP, you can be asking questions of Crowd Threat data in three steps.
Get your access
Get in touch and we'll set you up with your Crowd Threat connection details.
Add Crowd Threat to your AI tool
Add the Crowd Threat MCP server as a custom connector in your AI tool's settings.
Start asking
Ask about any place, time window or type of incident, and your AI will pull in verified reporting.
what we've done with it
Real questions we've put to the Crowd Threat MCP from our own AI tools, and what came back.
first field notes landing soon
We're logging the real questions we ask the Crowd Threat MCP and what it finds. Check back soon.
common questions
put verified ground truth into your ai
Connect your AI tool to the Crowd Threat MCP and start asking questions of our global reporter network's data.
Join our open OSINT community.
Open to anyone interested in OSINT. Some areas are reserved for verified Crowd Threat reporters.
The Spire
Pilgrim Street
Newcastle upon Tyne
NE1 2DS
United Kingdom