Log GDPR Article 22 Decisions w/ Groq & Discord Alerts

Log GDPR Article 22 decisions and trigger Groq-powered Discord alerts

This n8n workflow captures automated decision records from an n8n form, evaluates them against GDPR Article 22 criteria using Groq and configurable “effect” rules, and alerts your team in Discord when safeguards are needed—or when the lawful basis is missing or insufficient.

What this workflow does

• One-time setup: Run an initial setup to create four n8n Data Tables and load a predefined set of Article 22 “effect” rules into the rules table.
• Evaluate each decision: Receive a decision record from an n8n form submission, load the effect rules, and check whether the decision is solely automated and whether form answers/keyword matches indicate legally or similarly significant effects.
• Groq-assisted effect proposal: Send the decision text and rules to Groq via an LLM chain so it can propose additional effects only when it can provide verbatim quotes from the submitted decision details.
• Generate an Article 22 verdict + report: Produce an Article 22 verdict, including “INCOMPLETE” when there isn’t enough information, and create an evidence-backed report explaining which effects were counted and why.
• Persist findings: Insert the decision, verdict, and report into art22_decisions, and log each identified effect (or a “none” marker) into art22_effect_findings.
• Discord escalation + workflow completion: If safeguards are required or no valid lawful basis is recorded, post an alert with the report to Discord and show a dedicated completion page for the person logging the decision.
• Human review tracking: Accept a human review request from a second n8n form and initiate weekly reminders (as described in the workflow overview).

Use cases

• Audit automated decision flows (e.g., eligibility, profiling, or risk scoring) and document GDPR Article 22 assessments.
• Centralize evidence and effect reasoning with repeatable Groq + keyword/effect-rule checks.
• Ensure teams are notified in Discord when Article 22 safeguards are needed or when records are incomplete.

Technical details

• Core nodes: if, set, code, form, discord, data table
• Data persistence: art22_decisions and art22_effect_findings (via n8n Data Tables)
• LLM integration: Groq (LLM chain) to propose additional effects with verbatim quotes