Valkri Dashboard — Open-Source Geospatial Intelligence Dashboard

A real-time intelligence dashboard built around an interactive 3D globe, designed to aggregate, visualize, and analyze open-source data streams across geopolitical events, infrastructure, and threat landscapes.

Vision

Note: This project is in active early development. Features, tech stack choices, and architecture are all subject to change as the project evolves.

Valkri brings together disparate OSINT data sources into a single operational picture. The core interface is an OpenGlobus-powered 3D globe where users can explore live events spatially — from active conflict zones and shipping lanes to infrastructure outages and natural disasters — backed by a live news feed and structured reporting tools.

The initial focus is on the Russo-Ukrainian conflict as a proving ground, given the volume and variety of available data. The platform is designed to scale across any region or event type.

Core Features

3D Globe (OpenGlobus)

The primary interface is an interactive globe with layered data overlays. A 2D map toggle is planned for accessibility and performance on lower-end devices.

Data layers include:

• Shipping lanes — AIS vessel tracking and maritime route visualization
• Air travel routes — Flight path data and airspace activity
• Railway networks — Rail infrastructure and service maps
• Infrastructure announcements — Physical markers for deployments, construction, outages
• Down Detector integration — Service outage heatmaps and status
• Geotagged video — YouTube and other platforms, plotted by location
• Radiation mapping — Environmental monitoring station data
• Natural disasters — Earthquakes, weather events, geological disturbances
• Blockchain transactions — Geospatially-tagged on-chain activity
• ALPR reporting — License plate reader data (via Deflock.me and similar sources)
• Conflict tracking — Active wars, territorial control, frontline changes
• Threat actor monitoring — Inspired by platforms like Cognyte's external threat intelligence

Live News Feed

A multi-source aggregated feed pulling from:

• Twitter/X — Filtered for OSINT-relevant accounts and keywords
• Telegram — Channel monitoring for conflict and regional intel
• Traditional news — RSS and API-based ingestion from major outlets

Reporting Engine

Structured report creation with an emphasis on integrity:

• Templated and freeform intelligence reports
• Immutable report storage (blockchain-anchored hashing under consideration)
• Link analysis visualization inspired by Maltego's approach

Tech Stack

Frontend

• Framework: Next.js (TypeScript/React)
• Globe: OpenGlobus
• State management: Redux (TypeScript)
• Graph visualization: Cytoscape.js
• Mapping fallback: Mapbox GL JS

Backend

• API layer: Rust (high-performance HTTP/gRPC)
• Ingest services: Python — scrapers, API connectors, RSS consumers
• Processing services: Python — NLP, entity extraction, deduplication, geocoding

Data Layer

• Graph database: Neo4j or TigerGraph (entity relationships, link analysis)
• Vector search: Qdrant or pgvector (semantic search across reports and feeds)
• Message queue: Kafka or RabbitMQ (service-to-service communication)

Architecture

┌─────────────┐     ┌──────────────┐     ┌──────────────┐
│   Frontend  │◄───►│   Rust API   │◄───►│   Neo4j /    │
│  Next.js +  │     │   Layer      │     │  TigerGraph  │
│  OpenGlobus │     └──────┬───────┘     └──────────────┘
└─────────────┘            │
                     ┌─────┴──────┐
                     │   Kafka /  │
                     │  RabbitMQ  │
                     └─────┬──────┘
                     ┌─────┴──────────────────┐
                     │                        │
              ┌──────┴──────┐         ┌───────┴───────┐
              │   Ingest    │         │  Processing   │
              │  (Python)   │         │   (Python)    │
              │  Scrapers,  │         │  NLP, Entity  │
              │  APIs, RSS  │         │  Extraction   │
              └─────────────┘         └───────────────┘

Development Approach

Frontend-first. The strategy is to build a polished, fluid UI/UX before integrating backend services. Test data and mock APIs will stand in for live sources during early development, allowing the team to validate the user experience independently.

Key principles:

• Each data layer should be toggleable and performant on the globe
• Avoid click-heavy navigation patterns — information should surface contextually, not behind menus for each conflict zone
• Design for the intelligence lifecycle: collection → processing → analysis → dissemination

Getting Started

This is a Next.js project bootstrapped with create-next-app.

# Install dependencies
npm install

# Run the development server
npm run dev

Open http://localhost:3000 to view the app.

References and Inspiration

• CesiumJS Sandtable — Military Planning
• Maltego Link Analysis
• Cognyte External Threat Intelligence
• Deflock.me — ALPR Reporting
• Intelligence Lifecycle Thread

License

TBD

Contributing

TBD