hunting-domain-fronting-and-cdn-abuse
Hunts for domain fronting and CDN abuse by detecting TLS SNI versus HTTP Host header mismatches and high-reputation CDN domains carrying anomalous traffic in network telemetry. Activates for requests
SKILL.mdskills/hunting-domain-fronting-and-cdn-abuse/SKILL.md
GitHubHunting Domain Fronting and CDN Abuse
When to Use
- You have TLS/HTTP telemetry (Zeek ssl.log + http.log, proxy logs) and want to detect domain fronting, where the TLS SNI shows a benign CDN domain but the inner HTTP Host differs.
- You are hunting C2 hidden behind high-reputation CDNs.
Do not use SNI/Host mismatch as definitive — some legitimate setups differ. Corroborate with destination behavior and reputation.
Prerequisites
- Telemetry pairing TLS SNI with the inner HTTP Host (or proxy logs exposing both).
Workflow
Step 1: Detect SNI/Host mismatch
python scripts/analyst.py hunt flows.csv
Flags connections where the TLS SNI and HTTP Host headers resolve to different registered domains, especially when the SNI is a known CDN/fronting-capable domain.
Step 2: Surface CDN-hosted anomalies
Highlight high-reputation CDN SNIs carrying long-lived or beaconing flows.
Step 3: Confirm
Corroborate with beacon periodicity, JA3, and destination reputation; many CDN mismatches are benign.
Step 4: Operationalize
Write a detection for SNI/Host registered-domain mismatch on fronting-capable CDNs.
Validation
- Mismatch is computed on the registered domain, not the full hostname.
- Known CDN SNIs are recognized to prioritize fronting-capable cases.
- Findings are presented with corroboration needs, not as proof.
Pitfalls
- Legitimate multi-tenant CDN/edge configs producing benign mismatches.
- Encrypted SNI (ECH) hiding the SNI entirely — different hunt required.
- Treating any CDN traffic as suspicious (high false-positive rate).
References
- See
references/api-reference.mdfor the hunter. - ATT&CK T1090.004 and the TLS SNI RFC (linked in frontmatter).