Skip to content
UsecaseBoard
← All use cases
n8nbuilder

Evaluate a support agent with a separate correctness-and-helpfulness judge

At a glance

The template runs question-and-reference-answer cases through a support agent and uses a separate GPT-4 judge to score open-ended responses in n8n's evaluation view. This workflow is listed under builder, evals, support-agents, quality.

Agents / tools
n8n, GPT-4, n8n Evaluations

Building a tool for these workflows? See weekly rates.

Save ideas in this browser. Upvotes count across the board.

Prompt excerpt

Open the production agent path and configure the separate judge model. Add representative support questions and expected answers to the evaluation dataset. Run the evaluation, inspect score justifications alongside token use and latency, and manually review disagreements before changing the agent's prompt.
Explore all →
docsbuilder

Detect duplicate invoices and pause risky approvals in n8n

This n8n template extracts invoice PDFs with GPT-4o, checks invoice numbers and totals, assesses category and risk, and uses a wait-and-resume approval path before recording the final status.

n8nOpenAI GPT-4oGoogle Sheets+1
Source ↗