n8n Privacy Gate Workflow: Block Personal Data to Gmail/Telegram

Stop personal data from leaving your network—until this n8n Privacy Gate says it’s safe

This n8n Privacy Gate workflow scores every outbound payload for personal data before it reaches your partner APIs (including Gmail/Telegram destinations), then automatically blocks, allows, or holds the send. Each decision is logged to Google Sheets and the privacy officer is notified with the outcome.

What this workflow does

• Receives a gate request via webhook: A POST hits jev-c3-privacy-gate with job_id, record_ref, payload_text, and destination.
• Rejects incomplete requests early: It checks that payload_text and destination are present. If either is missing, it skips the model call, marks the request as UNUSABLE_GATE_REQUEST, prevents external sending, logs appropriately, and alerts ops by email.
• Classifies personal data with Jev: The workflow runs a Jev Classification score on the payload text.
• Applies decision bands:
  • ≥ 0.85: block the send (external_send_allowed = false)
  • ≤ 0.15: allow the send
  • Between: hold the send for the privacy officer
• Logs + notifies every outcome: Every decision writes a row to the “Privacy Gate” tab in Google Sheets first, then sends notifications by email and Telegram.
• Daily audit support: A 07:00 daily lane counts decisions and uses gpt-6-luna to draft the privacy audit note.

Use cases

• SaaS teams routing outbound messages and partner API calls that may contain customer details
• Automation engineers needing a repeatable, auditable privacy gate in n8n
• Privacy officers reviewing “hold” cases before sensitive data is forwarded

Technical details

• Platform: Runs on n8n
• Inputs/Triggers: webhook on jev-c3-privacy-gate
• Control logic: if, set
• Notifications: gmail and telegram
• Documentation/audit: Writes to Google Sheets (“Privacy Gate” tab) and summarizes the daily audit note (using gpt-6-luna)

Last updated: October 2026.