{"product_id":"ai-support-agent-regression-testing-in-n8n-google-sheets-groq-gmail","title":"AI Support Agent Regression Testing in n8n: Google Sheets, Groq \u0026 Gmail","description":"\u003ch3\u003eRegression-test your customer-support AI in n8n—using Google Sheets, Groq, and Gmail\u003c\/h3\u003e\n\u003cp\u003eThis n8n workflow runs an AI support agent against your Google Sheets test cases, scores the quality with Groq-hosted LLMs, logs everything back to Sheets, and automatically sends Gmail alerts when results regress or need QA review.\u003c\/p\u003e\n\n\u003ch3\u003eWhat this workflow does\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003eManual start:\u003c\/strong\u003e Designed to begin when you manually execute the workflow in n8n.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eLoads test scenarios:\u003c\/strong\u003e Reads rows from a Google Sheets “Test_Cases” tab, filters to \u003cstrong\u003eActive\u003c\/strong\u003e scenarios, and processes them in batches.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eGenerates support answers:\u003c\/strong\u003e For each test case, it constructs a customer-support prompt and generates an agent response using \u003cstrong\u003eGoogle Gemini\u003c\/strong\u003e.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eEvaluates response quality with Groq:\u003c\/strong\u003e Sends the question, \u003cstrong\u003eExpected_Facts\u003c\/strong\u003e, \u003cstrong\u003eExpected_Outcome\u003c\/strong\u003e, and the agent response to a \u003cstrong\u003eGroq\u003c\/strong\u003e LLM that returns scoring as JSON—covering relevance, reference accuracy, completeness, instruction compliance, and hallucination risk.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eLogs per-test results:\u003c\/strong\u003e Parses the evaluator JSON and appends detailed results (including the agent response and overall result) to an \u003cstrong\u003eEvaluation_Results\u003c\/strong\u003e sheet.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eCompares against a baseline:\u003c\/strong\u003e Aggregates the current run, selects the latest previously reviewed baseline from Google Sheets, and computes \u003cstrong\u003epass-rate regression\u003c\/strong\u003e and metric deltas.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eRoutes outcomes to QA via Gmail:\u003c\/strong\u003e Writes run-level metrics back to Google Sheets, then marks success or sends Gmail warning\/alert messages to the QA team.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eUse cases\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003eCatch customer-support regressions after prompt\/model changes in your AI agent.\u003c\/li\u003e\n  \u003cli\u003eContinuously monitor answer quality across a curated set of scenarios maintained in Google Sheets.\u003c\/li\u003e\n  \u003cli\u003eAutomate QA review workflows by alerting the team when baseline comparisons fail.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eTechnical details\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003eIntegrations:\u003c\/strong\u003e Google Sheets (Test_Cases, Evaluation_Results, Baseline_Metrics) and Gmail.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eLLMs:\u003c\/strong\u003e Google Gemini for agent responses; \u003cstrong\u003eGroq\u003c\/strong\u003e LLM for JSON evaluation scoring.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003en8n nodes\/logic:\u003c\/strong\u003e if, filter, switch, set, code, and gmail for control flow and reporting.\u003c\/li\u003e\n\u003c\/ul\u003e","brand":"N8N Commerce","offers":[{"title":"Default Title","offer_id":46086665601203,"sku":"N8N-19899","price":30.99,"currency_code":"GBP","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0749\/6279\/6723\/files\/CAPfFWWJOY1jd8FZ0QH1C_j0A3htEW.png?v=1790327581","url":"https:\/\/buyflowscripts.com\/products\/ai-support-agent-regression-testing-in-n8n-google-sheets-groq-gmail","provider":"N8N Commerce","version":"1.0","type":"link"}