Introduction
"Real-time global intelligence dashboard — AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface."
This is article #131 in the "One Open Source Project a Day" series. Today's project is worldmonitor — a real-time global intelligence dashboard that integrates AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking into a unified situational awareness interface.
73,260 Stars. Created January 2026. One of the fastest-growing open source projects on GitHub this year. One word in the project's Topics list says a lot about its positioning: palantir — an explicit declaration that this is the open-source answer to Palantir.
What You'll Learn
- worldmonitor's six core functional modules
- The dual map engine (3D globe + WebGL flat map) and the design logic behind 56 layer types
- The Country Instability Index (CII v8) scoring mechanism
- Local AI integration: full Ollama support, no API keys required
- MCP server, REST API, CLI, and multi-language SDK access patterns
- The design rationale behind 6 site variants (world / tech / finance / commodity / energy / happy)
Prerequisites
- Basic familiarity with web applications
- General interest in geopolitics, financial markets, or OSINT (open-source intelligence)
Project Background
Overview
worldmonitor is an intelligence platform that integrates multi-dimensional global data into a single interface. It's simultaneously a data backend for AI agents: it exposes an MCP server endpoint so any MCP-compatible AI tool can query real-time global intelligence data programmatically.
Author / Team
- Author: Elie Habib (koala73)
- Language: TypeScript
- License: AGPL-3.0-only
- Website: worldmonitor.app
Project Stats
- ⭐ GitHub Stars: 73,260+
- 🍴 Forks: 10,984+
- 📄 License: AGPL-3.0
- 📅 Created: 2026-01-08
Quick Start
git clone https://github.com/koala73/worldmonitor.git
cd worldmonitor
npm install
npm run dev
Open localhost:3000. No environment variables required to run — specific data sources need API credentials, but the core interface works out of the box.
Start a specific site variant:
npm run dev:tech # tech.worldmonitor.app
npm run dev:finance # finance.worldmonitor.app
npm run dev:commodity # commodity.worldmonitor.app
npm run dev:happy # happy.worldmonitor.app
npm run dev:energy # energy.worldmonitor.app
Or use the hosted versions directly, no installation needed:
- worldmonitor.app — general (geopolitics/military/disaster)
- tech.worldmonitor.app — tech intelligence
- finance.worldmonitor.app — financial markets
- commodity.worldmonitor.app — commodities
- energy.worldmonitor.app — energy markets
- happy.worldmonitor.app — positive news (a deliberate counterweight to the other five)
Six Core Modules
1. AI News Aggregation Engine
500+ curated feeds across 15 categories, AI-synthesized into briefs.
Key capabilities:
- Cross-stream correlation: converges military, economic, disaster, and escalation signals to automatically detect intersections across multiple data dimensions
- Local AI: run all AI synthesis through Ollama locally — zero cloud API keys required
- 25 languages with native-language feeds and RTL layout support
2. Dual Map Engine + 56 Layer Types
The most visually striking feature of worldmonitor:
3D globe mode (globe.gl + Three.js):
- Spherical Earth rendering with rotate/zoom
- Real-time event annotations overlaid on the globe
WebGL flat map mode (deck.gl + MapLibre GL):
- GPU-accelerated flat map rendering
- Better suited for analyzing regional event distributions
56 map layer types, including:
- Military activity hotspots
- Natural disasters (earthquakes, floods, wildfires)
- Infrastructure status (power grids, ports, airports)
- Live flight tracking (ADS-B data from Wingbits)
- Maritime/shipping routes
- Cyber attack source distribution
- Climate/environmental data
3. Country Instability Index (CII v8)
worldmonitor's proprietary Country Instability Index (CII), server-authoritative, currently at version 8:
- Covers 31 Tier-1 countries
- Aggregates multi-dimensional signals: political unrest, armed conflict, economic stress, social events, and more
- Scoring algorithm runs server-side; clients consume results rather than computing independently (ensuring everyone sees consistent, canonical scores)
- Gives users a quantified answer to "how unstable is this country right now?"
This is one of the features that separates worldmonitor from a plain news aggregator: it doesn't just surface news, it uses algorithms to transform information into quantified situational awareness metrics.
4. Finance Radar
- Real-time data from 29 stock exchanges
- Commodities (oil, gold, natural gas, etc.)
- Cryptocurrency markets
- 7-signal market composite: combines multi-dimensional financial signals into a single risk/opportunity score
The Finance variant (finance.worldmonitor.app) elevates financial data to the primary view, making it easier to cross-reference market movements with geopolitical events.
5. Native Desktop App (Tauri 2)
Native binaries for Windows, macOS (Apple Silicon / Intel), and Linux (AppImage), built on Tauri 2 (Rust + WebView):
- Single Tauri binary that switches between variants in-app
- Node.js sidecar handles API logic
Direct download links via the official API:
https://www.worldmonitor.app/api/download?platform=windows-exe
https://www.worldmonitor.app/api/download?platform=macos-arm64
https://www.worldmonitor.app/api/download?platform=linux-appimage
6. 65+ Data Source Integrations
External providers across:
- Geopolitical intelligence
- Financial market data
- Energy (oil/gas, renewables)
- Climate and environment
- Aviation (ADS-B live flight tracking)
- Cybersecurity events
- Military activity
- Infrastructure status
- News intelligence
500+ curated feeds, freshness monitoring across 35 source groups.
Technical Architecture
Frontend
Vanilla TypeScript + Vite
│
├── 3D globe: globe.gl + Three.js
├── Flat map: deck.gl + MapLibre GL
└── Browser-side AI: Transformers.js (no server needed)
Worth noting: pure TypeScript with no React/Vue/Angular, built with Vite. A relatively uncommon choice in today's frontend ecosystem, but one that gives direct control over performance.
AI/ML Layer
Local inference: Ollama (fully local, zero API keys)
Cloud inference: Groq / OpenRouter (optional)
Browser-side: Transformers.js (lightweight, no server)
API Contract Layer
An unusual architectural choice: Protocol Buffers for all API contracts:
- 281 proto files
- 35 service definitions
- sebuf HTTP annotations
Deployment Layer
60+ Vercel Edge Functions (serverless edge compute)
Railway relay (relay service)
Tauri (desktop app)
PWA (Progressive Web App)
Caching strategy: Redis (Upstash) + three-tier cache + CDN + Service Worker.
Programmatic Access: Built for AI Agents
worldmonitor explicitly positions itself as a data backend for AI agents, with multiple access paths:
MCP Server
Endpoint: https://worldmonitor.app/mcp
Protocol: Streamable HTTP MCP
Auth: X-WorldMonitor-Key header or OAuth
tools/list is public (no key needed to discover all available tools); tools/call requires an API key.
CLI
# No install needed — list all MCP tools (no key required)
npx worldmonitor tools
# Install global command (alias: wm)
npm install -g worldmonitor
# Query risk data with API key
worldmonitor risk IR --api-key wm_xxx
Multi-Language SDKs
| Language | Package | Install |
|---|---|---|
| Python | worldmonitor-sdk |
pip install worldmonitor-sdk |
| Ruby | worldmonitor |
gem install worldmonitor |
| Go | github.com/koala73/worldmonitor/sdk/go |
go get ... |
| JavaScript/TypeScript | worldmonitor |
npm i worldmonitor |
All SDKs are zero-dependency client libraries mirroring the CLI's capabilities.
Agent Discovery Files
https://worldmonitor.app/llms.txt
https://worldmonitor.app/.well-known/agent-skills/index.json
https://worldmonitor.app/.well-known/api-catalog
worldmonitor follows emerging agent-discovery conventions so AI tools can automatically discover and describe its capabilities without manual configuration.
The 6-Variant Design Logic
Generating 6 differently-positioned sites from a single codebase is a product design choice worth examining:
| Variant | URL | Primary lens |
|---|---|---|
| world (general) | worldmonitor.app | Geopolitics, military, disaster, global dynamics |
| tech | tech.worldmonitor.app | Tech industry, AI, cybersecurity |
| finance | finance.worldmonitor.app | Markets × geopolitics cross-analysis |
| commodity | commodity.worldmonitor.app | Commodities, supply chains, resource geopolitics |
| energy | energy.worldmonitor.app | Oil/gas, renewables, energy politics |
| happy | happy.worldmonitor.app | Positive news (deliberate counterweight to the above five) |
The existence of the "happy" variant signals self-awareness. The other five variants all amplify negative signals — conflict, risk, volatility. "happy" is an explicit hedge, and an honest acknowledgment that a global intelligence dashboard is almost by definition designed to make users anxious. Shipping a counterweight to that is an unusually thoughtful product decision.
Resources
Official Links
- 🌟 GitHub: koala73/worldmonitor
- 🌐 Website: worldmonitor.app
- 📖 Docs: worldmonitor.app/docs/documentation
- 🚀 Getting Started: worldmonitor.app/docs/getting-started
- 🏗️ Architecture: worldmonitor.app/docs/architecture
- 📦 npm CLI: npmjs.com/package/worldmonitor
- 💬 Discord: discord.gg/re63kWKxaz
Summary
worldmonitor does something that looks overambitious but turns out to be solidly executed: it open-sources a Palantir-class intelligence dashboard. 73k Stars accumulated in roughly three months signals a genuine unmet need — global situational awareness tools have historically been the exclusive domain of commercial, closed-source products.
Three design decisions worth remembering:
AI-agent-first design: The MCP server endpoint, llms.txt, agent-skills manifest, and multi-language SDKs aren't afterthought plugins — they're designed as first-class access layers alongside the web interface. worldmonitor isn't only a dashboard for humans; it's a data layer that AI agents can query for real-time global intelligence.
Local AI first: Ollama support means the entire AI feature chain can run fully on-device with no cloud API dependency. For sensitive geopolitical queries, this is a meaningful privacy guarantee.
Single codebase, multiple variants: Generating 6 differently-positioned sites from one codebase shows that the team thought clearly about audience segmentation rather than dumping all data into one overwhelming interface.
Server-authoritative CII scoring: Rather than letting each client compute its own instability index, the scoring algorithm runs server-side and clients consume the canonical result. This ensures consistency and prevents the fragmentation that comes from clients running divergent algorithm versions.
If you track geopolitics, need to correlate global events with market movements, or are building an AI agent pipeline that needs real-time global data — worldmonitor is currently the most feature-complete open-source option in this space.
Explore PrimeSkills — a curated marketplace of AI agents and skills, each validated against real enterprise workflows. No hype, just what actually works.
Visit my personal site for more insights and interesting products.
Top comments (0)