Score customer support AI responses with GPT‑4 judge metrics
Needs changesNeeds changes. This template uses 1 thing(s) that n8n 3.0 removes: AI Agent v1 (old agent modes). 1 breaking item(s) found. Static check, last run 2026-10-19.
What this workflow does
This AI workflow has 10 nodes and runs on Evaluation events (plus 1 other trigger). Along the way it connects Evaluation, OpenAI Chat Model and OpenAI.
How it runs
Trigger: On Evaluation events (plus 1 other trigger).
- Format eval input (Code): runs custom JavaScript code
- AI Agent: runs an AI agent, with model: OpenAI Chat Model
- Evaluating? (Evaluation): check if evaluating
- Judge - Score Response (OpenAI): uses OpenAI
- Evaluation: uses Evaluation
- Set Metrics (Evaluation): set metrics
Services it uses
Before you upgrade to n8n 3.0
1 thing in this workflow is removed in n8n 3.0. Fix it before you upgrade:
- AI Agent v1 (old agent modes) in AI Agent (v1.9): replace with Latest AI Agent node (Tools Agent). For SQL Agent, use a Postgres or MySQL tool with a recent AI Agent.
Rather not rewrite it? Many of these flows can be rebuilt in Make (free plan).
| Removed in 3.0 | Replace with |
|---|---|
| AI Agent v1 (old agent modes) AI Agent (v1.9) | Latest AI Agent node (Tools Agent). For SQL Agent, use a Postgres or MySQL tool with a recent AI Agent. |
Nodes used (8)
- Chat Trigger
@n8n/n8n-nodes-langchain.chatTrigger - Evaluation Trigger
n8n-nodes-base.evaluationTrigger - Code
n8n-nodes-base.code - AI Agent
@n8n/n8n-nodes-langchain.agent - OpenAI Chat Model
@n8n/n8n-nodes-langchain.lmChatOpenAi - Evaluation
n8n-nodes-base.evaluation - No Operation, do nothing
n8n-nodes-base.noOp - OpenAI
@n8n/n8n-nodes-langchain.openAi
From the author
Score open-ended AI responses with a judge model. This template shows how to evaluate a customer support agent using a separate LLM that rates each response on correctness and helpfulness, going beyond what exact match scoring can capture. What you'll do Open…