This n8n workflow is designed to streamline website visual regression testing using AI vision models, primarily Google Gemini. It automates the process of capturing webpage screenshots, comparing them to detect visual differences, and generating detailed reports of changes. The workflow is split into two main parts: Part A focuses on generating and storing base images for reference, while Part B performs the actual comparison to identify unwanted visual changes.

In Part A, the workflow retrieves a list of webpages from Google Sheets, captures baseline screenshots using Apify.com, and uploads these images to Google Drive, saving their references. This setup prepares the system for ongoing visual tests.

Part B is triggered either manually or via schedule, where it fetches the latest webpage URLs, generates new screenshots, and compares them against the stored base images. Using the AI Vision Model, it detects differences such as content shifts, color changes, or layout distortions. The results are filtered to identify discrepancies, which are then compiled into a report sent to Linear.app, helping teams quickly assess website updates or issues.

This automation is particularly useful for web developers, QA teams, and digital marketers who need to regularly verify website consistency and quickly spot unexpected changes. It offers a repeatable, scalable solution for maintaining webpage integrity over time with minimal manual intervention.

Node Count	>20 Nodes
Nodes Used	@n8n/n8n-nodes-langchain.chainLlm, @n8n/n8n-nodes-langchain.lmChatGoogleGemini, @n8n/n8n-nodes-langchain.outputParserStructured, aggregate, filter, googleDrive, googleSheets, httpRequest, linear, manualTrigger, merge, scheduleTrigger, set, splitInBatches, stickyNote, wait