Medical Webhook Triage: Claude + Slack Safety Alerts (n8n)

Instant medical-safety triage for webhook questions—powered by Claude and Slack

Medical Webhook Triage: Claude + Slack Safety Alerts (n8n) automatically receives medical information requests via webhook, uses Anthropic Claude to classify topic/urgency/complexity and detect safety signals, and then either sends a sourced email response or escalates instantly with Pharmacovigilance case creation and Slack alerts. Built for compliance-focused n8n workflows.

What this workflow does

• Receives requests through a POST webhook containing the question, requester details, and product name.
• Classifies with Claude: determines the topic, urgency, and complexity, and flags any potential adverse event or other safety signal. If classification/parsing fails, the workflow treats it as high-risk.
• Escalates safety concerns: when a safety signal is flagged (or parsing fails), it creates an urgent Pharmacovigilance case via an HTTP API and sends an immediate alert to the Drug Safety Slack channel.
• Drafts sourced answers when safe: otherwise, it searches an approved medical-information knowledge base via HTTP, checks coverage, and asks Claude to draft a reply strictly from retrieved sources with citations.
• Routes or escalates: if the request is auto-answerable (Standard complexity, Routine urgency, sufficient coverage, and a complete draft), it emails the sourced answer. If not, it creates an escalated Medical Information case via HTTP and alerts the appropriate team/channel on Slack.
• Logs outcomes for traceability and compliance.

Use cases

• Handling patient or clinician inquiries submitted to your SaaS form and routing them to Medical Information vs. Pharmacovigilance.
• Automatic detection of adverse event or safety signals to trigger immediate Drug Safety escalation.
• Reducing response time while ensuring answers are grounded in an approved knowledge base with citations.

Technical details (n8n)

• Nodes/blocks: if, set, code, merge, slack, switch
• Integrations: webhook (POST), HTTP API, Slack alerting, email sending for auto-answered cases
• Core logic: Claude-based classification + safety flagging, knowledge-base retrieval, citation-grounded response drafting, compliance logging