fix(antispam): harden spam guard with signed form tokens and proxy-safe IP extraction
All checks were successful
Build & Deploy / 🔍 Prepare (push) Successful in 19s
Build & Deploy / 🧪 QA (push) Successful in 1m17s
Build & Deploy / 🏗️ Build (push) Successful in 2m33s
Build & Deploy / 🚀 Deploy (push) Successful in 25s
Build & Deploy / 🧪 Post-Deploy Verification (push) Successful in 51s
Build & Deploy / 🔔 Notify (push) Successful in 2s
All checks were successful
Build & Deploy / 🔍 Prepare (push) Successful in 19s
Build & Deploy / 🧪 QA (push) Successful in 1m17s
Build & Deploy / 🏗️ Build (push) Successful in 2m33s
Build & Deploy / 🚀 Deploy (push) Successful in 25s
Build & Deploy / 🧪 Post-Deploy Verification (push) Successful in 51s
Build & Deploy / 🔔 Notify (push) Successful in 2s
- Rate limit now uses the last X-Forwarded-For hop (appended by Traefik) instead of the client-controlled first hop, which bots rotated freely - Time-trap anchored to server time via HMAC-signed form token (FORM_TOKEN_SECRET) instead of client-supplied form_loaded_at - Fix RequestQuoteForm regression: FormData was built from scratch without honeypot/token hidden inputs and inputs lacked name attributes, so every legitimate quote request was silently blocked as spam
This commit is contained in:
@@ -53,6 +53,9 @@ GATEKEEPER_PASSWORD=klz2026
|
||||
# Required for the Payload CMS AI Chat Agent
|
||||
MISTRAL_API_KEY=
|
||||
|
||||
# Secret used to sign contact-form anti-spam tokens (server-anchored time-trap)
|
||||
FORM_TOKEN_SECRET=
|
||||
|
||||
# ────────────────────────────────────────────────────────────────────────────
|
||||
# Payload Infrastructure (Dockerized)
|
||||
# ────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
Reference in New Issue
Block a user