Route grounded RAG support tickets with Claude, Pinecone, Zendesk, Slack
Ready for 3.0What this workflow does
This workflow has 17 nodes and runs when an HTTP POST request hits its webhook URL. Along the way it connects Zendesk, Slack and Google Sheets. It also calls api.voyageai.com and api.anthropic.com over HTTP.
How it runs
Trigger: When an HTTP POST request hits its webhook URL.
- Normalize Incoming Query (Code): runs custom JavaScript code
- Post to Voyage AI Embeddings (HTTP Request): calls api.voyageai.com (POST)
- Search Vectors in Pinecone (HTTP Request): calls an HTTP API (POST)
- Process Search Results (Code): runs custom JavaScript code
- Check Retrieval Confidence (If): checks a condition and branches into true / false
- Post to Claude for Grounded Answer (HTTP Request): calls api.anthropic.com (POST)
- Build Escalation Package (Code): runs custom JavaScript code
- Validate AI Citations (Code): runs custom JavaScript code
- Create Escalation Ticket in Zendesk (Zendesk): uses Zendesk
- Check Escalation Need (If): checks a condition and branches into true / false
…plus 6 more steps. See the full node list below.
Services it uses
HTTP calls to: api.voyageai.com, api.anthropic.com
Before you upgrade to n8n 3.0
Nothing in this workflow is removed in n8n 3.0, so it should import and run after the upgrade.
Nodes used (9)
- Webhook
n8n-nodes-base.webhook - Code
n8n-nodes-base.code - HTTP Request
n8n-nodes-base.httpRequest - If
n8n-nodes-base.if - Zendesk
n8n-nodes-base.zendesk - Slack
n8n-nodes-base.slack - Respond to Webhook
n8n-nodes-base.respondToWebhook - Merge
n8n-nodes-base.merge - Google Sheets
n8n-nodes-base.googleSheets
From the author
Quick overview This workflow receives support questions via a webhook, retrieves relevant context from Pinecone using Voyage AI embeddings, and uses Anthropic Claude to draft a grounded, cited answer. It auto-replies when confidence is high, otherwise…