World Monitor: Open-Source Global Intelligence Dashboard with AI-Powered Signal Correlation

By Prahlad Menon 4 min read

The gap between what institutional intelligence desks see and what the rest of us can access has been shrinking fast. World Monitor (GitHub) might be the most aggressive attempt yet to close it entirely — an open-source global intelligence dashboard that pulls 500+ curated news feeds across 15 categories, synthesizes them with local AI, and layers everything onto dual map engines with 45 data overlays.

It’s a lot. Let’s break down what actually matters.

What You Get

World Monitor’s core pitch is cross-stream correlation. Instead of checking military trackers, economic dashboards, and disaster feeds separately, it tries to connect signals across domains — a troop movement near a shipping lane, an energy price spike coinciding with infrastructure disruption, an escalation signal converging with market volatility. This is the kind of analysis that previously required a team and a Bloomberg terminal.

The dual map engine gives you two views: a 3D globe built on globe.gl and Three.js for the cinematic overview, and a WebGL flat map using deck.gl and MapLibre GL for serious analytical work. Forty-five data layers cover everything from conflict zones to flight paths (sourced from Wingbits ADS-B data) to cyber incidents.

The Country Intelligence Index scores nations across 12 signal categories into a composite risk rating. The finance radar tracks 92 stock exchanges, commodities, and crypto through a 7-signal market composite. These aren’t just widgets — they feed into the correlation engine.

Everything runs on local AI via Ollama. No API keys, no usage caps, no data leaving your machine. The AI synthesizes raw feeds into structured briefs, which is critical when you’re processing 500+ sources.

Other details worth noting:

  • 5 site variants from a single codebase — world, tech, finance, commodity, and (charmingly) “happy”
  • Native desktop app via Tauri 2 for macOS, Windows, and Linux
  • 21 languages with native-language feeds and RTL support
  • 65+ external data sources spanning geopolitics, finance, energy, climate, aviation, cyber, military, and infrastructure
  • Built with vanilla TypeScript, Vite, 92 Protocol Buffer definitions, Vercel Edge Functions, and Redis caching
  • Licensed AGPL-3.0 for non-commercial use

How It Compares

We’ve covered two other tools in this space — ShadowBroker and ThinkCreate Intel. World Monitor sits in interesting territory relative to both.

ShadowBroker is geospatial-first. It’s a map with 15+ live feeds — aircraft, ships, satellites, GPS jamming, active conflicts. Raw OSINT plotted on a map. It answers “where is something happening right now?” brilliantly and doesn’t try to do more. That focus is its strength.

ThinkCreate Intel goes the other direction — AI threat scoring with LVL 1-10 ratings, FLASH/PRIORITY/ROUTINE classification, StockScout trade signals, and Telegram alerts. It’s the curated production layer that tells you what matters and how urgently.

World Monitor occupies the middle ground, but at much larger scale. It has ShadowBroker’s mapping DNA (dual engines, 45 layers vs. 15+ feeds) plus its own AI synthesis layer (Ollama-powered briefs instead of ThinkCreate’s cloud-based scoring). Think of it as ShadowBroker evolved into a full platform — more feeds, more analysis, more integration points.

The key differentiator is the cross-stream correlation engine. ShadowBroker shows you military movements. ThinkCreate scores threat levels. World Monitor tries to connect military movements with economic signals with disaster events and surface the convergence. Whether it does this reliably depends on your Ollama model and the quality of incoming data, but the architecture is there.

The Practical Take

If you want raw geospatial OSINT with zero setup, ShadowBroker remains the fastest path. If you want curated alerts pushed to Telegram with trade signals, ThinkCreate Intel is purpose-built for that workflow.

World Monitor is for the person who wants to run their own intelligence desk. The local AI requirement means you need decent hardware. The 500+ feed ingestion means you need bandwidth. But if you’re willing to invest the setup time, you get something that doesn’t exist elsewhere in the open-source world — a self-hosted, multi-domain intelligence platform with signal correlation across geopolitics, finance, and infrastructure.

The GitHub repo is active. The AGPL-3.0 license keeps it open for personal and research use. If you’ve outgrown single-domain dashboards and want convergence analysis without institutional pricing, this is worth your afternoon.