L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

hunting-for-domain-fronting-c2-traffic

hunting-for-domain-fronting-c2-traffic

Detect domain fronting C2 traffic by analyzing SNI vs HTTP Host header mismatches in proxy logs and TLS certificate discrepancies using pyOpenSSL for certificate inspection

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: hunting-for-domain-fronting-c2-traffic description: Detect domain fronting C2 traffic by analyzing SNI vs HTTP Host header mismatches in proxy logs and TLS certificate discrepancies using pyOpenSSL for certificate inspection domain: cybersecurity subdomain: threat-hunting tags:

  • domain-fronting
  • c2-detection
  • tls-inspection
  • proxy-logs
  • pyopenssl
  • threat-hunting
  • network-security version: "1.0" author: mahipal license: Apache-2.0

Hunting for Domain Fronting C2 Traffic

Overview

Domain fronting (MITRE ATT&CK T1090.004) is a technique where attackers use different domain names in the TLS SNI field and the HTTP Host header to disguise C2 traffic behind legitimate CDN-hosted domains. This skill detects domain fronting by parsing proxy/web gateway logs for SNI-Host header mismatches, analyzing TLS certificates for CDN provider identification, flagging connections where the SNI points to a high-reputation domain but the Host header targets an attacker-controlled domain, and correlating with known CDN provider IP ranges.

Prerequisites

  • Web proxy or secure web gateway logs with SNI and Host header fields
  • Python 3.8+ with pyOpenSSL and cryptography libraries
  • TLS inspection enabled on proxy for Host header visibility
  • CDN provider IP range lists (CloudFront, Azure CDN, Cloudflare)

Steps

  1. Parse proxy logs for connections with both SNI and Host header fields
  2. Compare SNI domain against HTTP Host header for mismatches
  3. Extract TLS certificate Subject and SAN fields using pyOpenSSL
  4. Identify CDN-hosted connections via certificate issuer and IP ranges
  5. Flag high-confidence domain fronting where SNI and Host differ on CDN IPs
  6. Score alerts based on domain reputation differential
  7. Generate detection report with network flow context

Expected Output

JSON report containing detected domain fronting indicators with SNI-Host pairs, certificate details, CDN provider identification, confidence scores, and MITRE ATT&CK technique mapping.

Skills enregistrées