{"product_id":"n8n-ci-webhook-gemini-prompt-grading-regression-checks","title":"n8n CI Webhook: Gemini Prompt Grading \u0026 Regression Checks","description":"\u003ch3\u003eRun Gemini prompt tests in your CI—automatically grade outputs, catch regressions, and fail the build when quality drops\u003c\/h3\u003e\n\u003cp\u003eThis \u003cstrong\u003en8n CI Webhook: Gemini Prompt Grading \u0026amp; Regression Checks\u003c\/strong\u003e exposes a secured webhook your CI pipeline can call to execute a prompt test suite against \u003cstrong\u003eGoogle Gemini\u003c\/strong\u003e. It grades each output against your rules, detects regressions versus the previous run, and returns an HTTP status code (pass or fail) to gate deployments.\u003c\/p\u003e\n\n\u003ch3\u003eWhat this workflow does\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003eReceives a CI POST request\u003c\/strong\u003e to a webhook secured with \u003cstrong\u003eheader authentication\u003c\/strong\u003e.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eValidates and expands the payload\u003c\/strong\u003e into individual test cases (including optional repeated runs per case) by filling the prompt with each input.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eExecutes the prompt-under-test\u003c\/strong\u003e for each case using \u003cstrong\u003eGoogle Gemini (PaLM)\u003c\/strong\u003e.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eGrading pass with a separate Gemini pass\u003c\/strong\u003e that strictly evaluates each output against the provided rules, returning structured results (pass\/fail, severity, broken rules).\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eAggregates into a scorecard\u003c\/strong\u003e, flags unstable inputs across repeated runs, and \u003cstrong\u003ecompares per-input verdicts\u003c\/strong\u003e against the last stored run to detect regressions.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eReturns a build gate response\u003c\/strong\u003e: HTTP \u003cstrong\u003e200\u003c\/strong\u003e with the scorecard when the pass-rate threshold is met and \u003cem\u003eno regressions\u003c\/em\u003e are found; otherwise HTTP \u003cstrong\u003e422\u003c\/strong\u003e with a failure report plus a Gemini-generated explanation and rewritten prompt.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eUse cases\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003ePrevent prompt regressions by failing CI when Gemini outputs no longer match grading rules.\u003c\/li\u003e\n  \u003cli\u003eContinuously validate prompt changes before deploying a SaaS feature that relies on Gemini.\u003c\/li\u003e\n  \u003cli\u003eDetect flaky\/unstable prompt behavior via repeated runs per input and severity-based broken rule reporting.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eTechnical details\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003en8n nodes\/workflow\u003c\/strong\u003e: if, code, webhook, sticky note, split in batches, respond to webhook.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eIntegrations\u003c\/strong\u003e: Google Gemini (PaLM) for both the model-under-test and grading\/rewrite steps.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eSetup requirement\u003c\/strong\u003e: activate the workflow so n8n \u003cstrong\u003estatic data persists between runs\u003c\/strong\u003e for regression comparison.\u003c\/li\u003e\n\u003c\/ul\u003e","brand":"N8N Commerce","offers":[{"title":"Default Title","offer_id":45824201162931,"sku":"N8N-18180","price":28.99,"currency_code":"GBP","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0749\/6279\/6723\/files\/fxegMj_koDo3UdRWQzoMz_VHGufwEB.png?v=1786611950","url":"https:\/\/buyflowscripts.com\/products\/n8n-ci-webhook-gemini-prompt-grading-regression-checks","provider":"N8N Commerce","version":"1.0","type":"link"}