Logo Lanfrica

RazForge/zebebgna

Domaine:

digital infrastructure

Type de record:

softwaretools
Créateur:
Raz
Hôte:
Guard - Defensive verification of Ethiopian bank receipts. Detects phishing lookalikes, validates TLS certificates, reconciles amounts, and fuses threat signals into a clear risk verdict. ZEBEGNA 🇿🇪 Guard — Defensive Verification of Ethiopian Bank Receipts Secure · Verified · Trusted --- ## What is Zebebgna? **Zebebgna** (Amharic: ዘበኛ — "guard") is a cybersecurity toolkit that verifies the authenticity of Ethiopian bank receipts. It extracts receipt data from **six major banks**, then puts that data on trial through a multi-layered verification pipeline. In 2025, Ethiopian bank fraud via fake receipts surged. Scammers create pixel-perfect copies of bank receipts or host them on phishing domains that look like `apps.cbe.com.et`. Victims lose millions of Birr trusting receipts that were never real. **Zebebgna catches what visual inspection misses.** --- ## How it works ``` Receipt Input (URL / PDF / Screenshot / Text) │ ▼ ┌─────────────────────────────────────────────────────────┐ │ EXTRACTION LAYER │ │ CBE · Dashen · Awash · BOA · Zemen · Telebirr │ │ PDF regex · HTML scraping · Headless Chrome · OCR │ └─────────────────────────┬───────────────────────────────┘ │ ▼ ┌─────────────────────────────────────────────────────────┐ │ VERIFICATION LAYER │ │ Phishing detection · TLS audit · Amount reconciliation │ │ Reference validation · Security headers · DNS checks │ └─────────────────────────┬───────────────────────────────┘ │ ▼ ┌─────────────────────────────────────────────────────────┐ │ FUSION ENGINE │ │ Correlates weak signals into attack scenarios │ │ Produces a deterministic 0–100 risk score │ └─────────────────────────┬───────────────────────────────┘ │ ▼ ┌─────────────────┐ │ VERDICT │ │ ✅ Good │ │ ⚠️ Needs Review│ │ ❌ Problem │ └─────────────────┘ ``` --- ## Features ### Extraction - **6 banks supported** — CBE, Dashen, Awash, Bank of Abyssinia, Zemen, Telebirr - **Multiple input formats** — URLs, PDFs, images (OCR), QR codes, pasted te …