Skip to content

GhostExodus OSINT Platform v1.0.0

Choose a tag to compare

View all tags
@Dezirae-Stark Dezirae-Stark released this 12 Apr 07:49
v1.0.0
4bd7d26

GhostExodus OSINT Platform v1.0.0

Counter-extremism intelligence monitoring suite for UK law enforcement support.

All data stays on device. No cloud dependencies. Fully local AI.

System Requirements

Component Requirement
OS Windows 10 build 17763+ or Windows 11 (64-bit only)
RAM 16 GB minimum
GPU NVIDIA GPU with 8 GB+ VRAM recommended (Ollama)
Storage 20 GB free

Must install first:

Building the Windows Installer

The installer must be built on a Windows machine:

  1. Install Python 3.11+, Node.js 18+, Inno Setup 6
  2. Clone: git clone https://github.com/Dezirae-Stark/CT-OSINT-AI-Tools.git
  3. Run: build_installer.bat
  4. Installer output: installer\output\GhostExodus_Setup_v1.0.0.exe

The build script automates: React frontend build → Python virtualenv → pip install → PyInstaller bundle → Inno Setup compilation.

Running From Source (Windows)

python -m venv venv && call venv\Scripts\activate
pip install -r requirements.txt
cd frontend && npm install && npm run build && cd ..
xcopy /E /Y /I frontend\dist backend\static
copy .env.example .env
:: Edit .env with TELEGRAM_API_ID, TELEGRAM_API_HASH, TELEGRAM_PHONE, JWT_SECRET
cd backend
uvicorn main:app --host 127.0.0.1 --port 8000

Open http://localhost:8000 — first-run wizard creates the admin account.

Platform Features

  • Real-time Telegram monitoring via Telethon MTProto with live WebSocket dashboard feed
  • 6-category keyword threat engine (Operational Planning, Recruitment, Financing, Propaganda, UK-Specific, Incitement) + custom regex alert rules
  • Local LLM classification — llama3.1:8b classifies MEDIUM+ severity messages (threat category, TTPs, UK relevance, urgency)
  • Semantic search & RAG — nomic-embed-text embeddings in ChromaDB, LlamaIndex pipeline for natural-language intelligence queries
  • Entity correlation graph — regex + LLM extraction of handles, phones, emails, crypto addresses, onion links; co-occurrence graph in React Flow
  • Stylometry — writing style fingerprinting with cosine similarity author comparison
  • SHA-256 evidence archive with per-item integrity verification and case bundle export (ZIP with manifest, chain-of-custody log, standalone Python verifier)
  • INTELREPORT PDF — 9-section formal report with classification header, LLM-generated executive summary, TTP assessment, evidence reference list
  • RBAC — ADMIN / ANALYST / VIEWER roles enforced on every API endpoint
  • Full audit trail — every user action logged to immutable audit table

Full Documentation

See README.md for the complete setup guide, configuration reference, UI walkthrough, API documentation, and architecture overview.

SENSITIVE — LAW ENFORCEMENT USE ONLY
Built for local deployment. No cloud dependencies. All data stays on device.
This platform is intended for use by authorised personnel only.
Assets 2
Loading