How it works

44 sources. 7 domains. One honest score.

No jargon. No black box. In three steps you'll know everything we do with the data — and everything we don't.

1 We watch

We listen to 44 verified feeds — every hour.

Real-time monitoring across health authorities (WHO, CDC, ECDC, ProMED, HealthMap), conflict databases (ACLED, GDELT), disaster systems (GDACS, USGS, NASA FIRMS), internet observatories (IODA, Cloudflare Radar), economic indicators (IMF, World Bank), and structural risk indices (UN INFORM).

Every event we show is sourced from one of these. No anonymous tipsters. No "industry insiders." No proprietary data we won't name.

WHO
CDC
ECDC
ACLED
GDELT
GDACS
USGS
NASA FIRMS
IODA
NetBlocks
IMF
World Bank
INFORM
ProMED
HealthMap
ReliefWeb
Wikipedia
Open-er-API
FAO
PAHO
Africa CDC
+ 23 more
Refreshed hourly · source-tagged on every record
2 We compute

The score shows its work.

For every country, the engine builds a composite score across 7 domains — health, conflict, civil unrest, transport, border, infrastructure, climate — then adjusts it honestly: amplified in structurally fragile countries (INFORM), damped in resilient ones, never declared "covert" based on a closed regime's silence alone (Bayesian discount).

Every score is traceable to its inputs. Click any flag — see exactly which signals contributed and where they came from. See the math →

Conflict
4.4
Civil unrest
2.7
Infrastructure
4.0
Health
0.9
Currency Δ30d
−20%
INFORM ×
1.6
4.87 / 5
Severe
3 You see

You get alerts that explain themselves.

Email to your security focal point, Telegram channel for the team, webhooks for your systems, or a clean dashboard for daily review — every alert lands with the score, the domain breakdown, and direct links to the underlying sources.

No "trust us" mode. Each notification is auditable.

Email alerts
To your security/safety focal point — the moment a domain crosses your threshold.
Telegram bot
@vigilocc_bot — for distributed teams who already live there.
Webhooks & REST API
HMAC-signed events to your stack — alerts, scores, raw events.
Dashboard & live globe
For situation rooms, briefings, and "I'll check before sending the team in" moments.

The 7 risk domains, in plain language.

No "indices," no acronyms. What we actually mean by each one.

Health
Outbreaks, epidemics, food safety recalls, wastewater anomalies, vaccine campaigns.
Conflict
Armed clashes, kinetic strikes, military operations. ACLED + GDELT sourced.
Civil unrest
Protests, demonstrations, riots, clusters of localized social pressure.
Disasters & climate
Earthquakes, floods, cyclones, wildfires, drought, climate anomalies.
Blackout
Internet shutdowns, telecom outages, power grid failures, cable cuts.
Border & transport
Entry restrictions, airspace closures, port disruptions, evacuation routes.
Food security
Acute insecurity, price shocks, FAO indices, supply-chain stress.
Macro
GDP, debt, unemployment, central-bank rates — lagging context, explainability primary.

Honest note on depth. All seven domains are scored for every country, but coverage density varies — conflict, health, climate and civil unrest are the deepest. Infrastructure, transport and border populate as events occur (a border closure or airport strike is a rarer event than a protest). We'd rather show you a sparse-but-real signal than pad a number.

Why "show your work" matters.

If you can't see why a system flagged a country, you can't defend the decision you make based on it. We took the opposite bet.

Typical vendor
Black-box score
  • "Our model says it's elevated. Trust us."
  • Methodology proprietary
  • Sources aggregated, not named
  • Can't audit a single decision
  • Hard to explain a wrong call
Vigilo
Source-traceable score
  • Every score links back to its raw signals
  • Methodology published at /methodology
  • Every source named — 44 of them
  • Open ingestion-batch IDs & provenance
  • Wrong calls are visible, fixable, & logged

Try it on your regions.

Open the live globe. Click any country. See the breakdown for yourself.