Skip to content

taniav1311/HECTOR

Repository files navigation

HECTOR – Explainable Port Risk Assessment System

HECTOR is a cybersecurity tool that converts raw Nmap scan results into prioritized, explainable risk insights.

Unlike traditional scanners that only list open ports, HECTOR identifies: what is risky, why it is risky, and what should be prioritized for remediation.


Core Idea

HECTOR uses a hybrid risk scoring model combining:

  • Heuristic risk (H) — domain knowledge of services
  • CVSS score (C) — real-time CVE data from NVD
  • Port weight (W) — contextual exposure relevance

Final Score = (0.4H + 0.6C) × W (with fallback logic when data is unavailable)

The primary contribution of this project is the design and implementation of an explainable hybrid risk-scoring engine.

The model combines:

  • Heuristic service risk knowledge (domain-specific rules)
  • CVSS-based vulnerability intelligence (NVD integration)
  • Port-level contextual weighting (practical exposure sensitivity)

These components are fused into a single normalized risk score, which is further mapped to operational severity levels (LOW, MEDIUM, HIGH, CRITICAL).

This approach ensures that the system:

  • Accounts for real-world context, not just theoretical severity
  • Maintains interpretability, with full visibility into scoring factors
  • Supports prioritization, enabling efficient remediation decisions

System Architecture

The system follows a modular pipeline:

image

Output

For each open port:

  • Risk score (0–10)
  • Severity (LOW → CRITICAL)
  • CVE mapping
  • Attack classification
  • Full scoring breakdown (transparent and auditable)
Screenshot 2026-04-30 120518 Screenshot 2026-04-30 121416 Screenshot 2026-04-29 192953 Screenshot 2026-04-29 193016 Screenshot 2026-04-29 193136 Screenshot 2026-04-29 193408

Key Features

Hybrid Risk Scoring

  • Combines heuristics, CVSS, and contextual weighting
  • Produces more realistic prioritization than single-source models

Explainability by Design

  • Complete visibility into scoring components
  • No black-box logic; fully reproducible results

CVE Intelligence Integration

  • Live NVD integration
  • Service-to-vulnerability mapping
  • Contextual enrichment of scan findings

Temporal Risk Tracking

  • Snapshot-based scan history
  • Tracks evolution of risk over time

Practical Scanning

  • Nmap-based scanning engine
  • Supports IPs, domains, and URLs
  • Input normalization with single-target enforcement

Technology Stack

Layer Technology Rationale
Frontend React Structured, modular UI for analytical workflows
Backend Flask (Python) Lightweight API layer for orchestration
Scanning Nmap Industry-standard network discovery tool
Risk Engine Python Flexibility for custom scoring logic
Data Source NVD (CVE) Authoritative vulnerability intelligence
Storage Local persistence Efficient snapshot-based history tracking

Differentiation

  • Hybrid scoring model (heuristics + CVE + context)
  • Fully explainable risk computation
  • Temporal tracking of exposure

Impact

  • Converts scan results into actionable priorities
  • Reduces manual analysis overhead
  • Enables consistent and auditable risk assessment
  • Supports continuous monitoring of security posture

Conclusion

HECTOR is an explainable vulnerability triage system that integrates scanning, enrichment, and scoring into a single pipeline.

Its value lies in interpreting and prioritizing exposure with clarity, context, and transparency.


TL;DR

HECTOR does not just detect vulnerabilities — it prioritizes and explains them.

About

Explainable port scanning with CVE-based risk assessment

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors