Advanced Intelligence and Model Context Protocol (MCP) ecosystem for OSINT, data analysis, and automated operations.
Intel-MCP is a comprehensive suite designed to bridge the gap between raw intelligence data and AI-driven analysis. It provides a robust backend for data collection and processing, a modern web interface for visualization, and an MCP server that grants AI agents (like Cursor or Claude) typed, safe access to intelligence tools.
The project is divided into three primary components:
- Backend (
/backend): A high-performance Python FastAPI service. Handles data orchestration, OSINT search, PDF generation, and background tasks using Celery/Redis. - Frontend (
/frontend): A modern Next.js web application. Features interactive dashboards, ECharts visualizations, Leaflet maps, and real-time data monitoring. - MCP Server (
/mcp): A Model Context Protocol implementation in TypeScript. Exposes backend capabilities as tools for AI hosts, enabling features likeintel.search,intel.digest, and automatedops.mcp_diag.
- Advanced OSINT: Integrated search capabilities with multi-source intelligence gathering.
- AI Integration: Native MCP support for Cursor and Claude Desktop.
- Data Visualization: Real-time dashboards, geographic mapping, and relationship graphing.
- Automated Reporting: Generate PDF/CSV/JSON digests and incident briefs.
- Operational Safety: Dry-run modes by default, audit logging, and strict policy enforcement.
- Node.js (v18+)
- Python 3.10+
- Docker & Docker Compose (for Redis/PostgreSQL)
- PNPM (for MCP and Frontend)
cd backend
python -m venv venv
source venv/bin/activate # or venv\Scripts\activate on Windows
pip install -e .
cp .env.example .env
# Configure your .env and start the server
uvicorn app.main:app --reload --port 8080cd frontend
pnpm install
pnpm devcd mcp
pnpm install
pnpm build
# Follow instructions in mcp/README.md for Cursor/Claude configurationIntel-MCP is built with operational security in mind:
- Audit Logging: Every tool call and system action is logged.
- Policy Control: Granular control via
policy.json. - Dry-run Protection: Destructive or high-risk actions require explicit confirmation.
Developed for high-stakes intelligence and operational environments.