⚡️ tensionr

Real-time Global Intelligence & Narrative Resonance Engine.

tensionr is a high-fidelity intelligence dashboard designed to monitor global instability, narrative shifts, and strategic asset movements in real-time. By synthesizing multi-domain signals (from GDELT news flows to ADS-B aerial telemetry) tensionr provides a unified "Geopolitical Tactical Picture" for rapid situational awareness.

🛰️ Core Capabilities

• LLM-Powered SITREP & Insight: Tactical bulletins summarized via DistilBART, plus a cross-domain "Strategic Insight" correlation synthesized via Mistral-7B-Instruct.
• Geopolitical Time Machine: A "Static Data Lake" of daily archives, allowing users to select past dates and visualize historical geopolitical shifts.
• Narrative Resonance Mapping: Uses NLP (GoEmotions) to extract granular emotional undertones (Fear, Anger, Sadness, etc.) from global news nodes.
• Strategic Aerial Telemetry: Real-time tracking of military assets with automated anomaly detection for strategic airframes.
• Multi-Domain GTI: A composite algorithmic score quantifying global instability based on news sentiment, market volatility (VIX, Gold), and tactical telemetry.
• Resilient Data Pipeline: Parallelized RSS fetching and smart API fallbacks ensuring continuous intelligence even under rate-limiting.

🎯 Who is it for?

• OSINT Analysts: Rapidly aggregate and filter unverified raw chatter and verified news nodes.
• Risk Managers: Monitor the GTI to assess corporate or asset exposure in volatile regions.
• Geopolitical Researchers: Visualize how narratives evolve and spread across different domains and languages.
• Data Enthusiasts: A showcase of real-time data orchestration using Python, JS, and automated CI/CD pipelines.

🛠️ Tech Stack

• Backend: Python (Data harvesting, GDELT integration, NLP sentiment analysis).
• Frontend: Vanilla JS / Bootstrap 5 / Leaflet.js (Tactical mapping and hardware-accelerated UI).
• Orchestration: GitHub Actions (automated telemetry sync every hour).
• Environments: Managed via uv with a committed uv.lock for reproducible builds.

🚀 Quick Start

1. Clone the repository:

   git clone https://github.com/exdsgift/tensionr.git
   cd tensionr

2. Setup environment:

   uv sync

3. Launch the Engine: Run the data pipeline to populate the dashboard (set HF_TOKEN in .env to enable the ML stages):

   uv run tensionr

   Then serve the dashboard locally (the app fetches data/*.json, so it needs an HTTP server, not file://):

   python -m http.server 8123

   and open http://localhost:8123.

4. Run the tests:

   uv run pytest

---

tensionr © 2026 // Engineered for the next generation of global observers.