Visual regression testing with Apify and AI Vision Model
This n8n workflow is a proof-of-concept template exploring how we might work with multimodal LLMs and their multi-image analysis capabilities. In this demo, we compare 2 screenshots of a webpage taken at different timestamps and pass both to our multimodal LLM for a visual comparison of differences. Handling multiple binary inputs (ie. images) in an AI request is supported by n8n's basic LLM node. How it works This template is intended to run as 2 parts: first to generate the base screenshots and next to run the visual regression test which captures fresh screenshots. Starting with a list of webpages captured in a Google sheet, base screenshots are captured for each using a external web scraping service called Apify.com (I prefer Apify but feel free to use whichever web scraping service available to you) These base screenshots are uploaded to Google Drive and will be referenced later when we run our testing. Phase 2 of the workflow, we'll use a scheduled trigger to fire sometime in the future which will reuse our web scraping service to generate fresh screenshots of our desired webpages. Next, re-download our base screenshots in parallel and with both old and new captures, we'll pass these to our LLM node. In the LLM node's options, we'll define 2 "user message" inputs with the type of binary (data) for our images. Finally, we'll prompt our LLM with our testing criteria and capture the regressions detected. Note, results will vary depending on which LLM you use. A final report can be generated using the LLM's output and is uploaded to Linear. Requirements Apify.com API key for web screenshotting service Google Drive and Sheets access to store list of webpages and captures Customising this workflow Have your own preferred web screenshotting service? Feel free to swap out Apify with your service of choice. If the web screenshot is too large, it may prove difficult for the LLM to spot differences with precision. Try splitting up captures into smaller images instead.
By JimleukGenerate AI images in bulk with Freepik, Google Sheets & Drive
This n8n workflow automates bulk AI image generation using Freepik's Text-to-Image API. It reads prompts from a Google Sheet, generates multiple variations of each image using Freepik's AI, and automatically uploads the results to Google Drive with organized file names. This is perfect for content creators, marketers, or designers who need to generate multiple AI images in bulk and store them systematically. Key Features: Bulk image generation from Google Sheets prompts Multiple variations per prompt (configurable duplicates) Automatic file naming and organization Direct upload to Google Drive Batch processing for efficient API usage Freepik AI-powered image generation Step-by-Step Implementation Guide Prerequisites Before setting up this workflow, you'll need: n8n instance (cloud or self-hosted) Freepik API account with Text-to-Image access Google account with access to Sheets and Drive Google Sheet with your prompts Step 1: Set Up Freepik API Credentials Go to Freepik API Developer Portal Create an account or sign in Navigate to your API dashboard Generate an API key for Text-to-Image service Copy the API key and save it securely In n8n, go to Credentials → Add Credential → HTTP Header Auth Configure as follows: Name: "Header Auth account" Header Name: x-freepik-api-key Header Value: Your Freepik API key Step 2: Set Up Google Credentials Google Sheets Access: Go to Google Cloud Console Create a new project or select existing one Enable Google Sheets API Create OAuth2 credentials In n8n, go to Credentials → Add Credential → Google Sheets OAuth2 API Enter your OAuth2 credentials and authorize with spreadsheets.readonly scope Google Drive Access: In Google Cloud Console, enable Google Drive API In n8n, go to Credentials → Add Credential → Google Drive OAuth2 API Enter your OAuth2 credentials and authorize Step 3: Create Your Google Sheet Create a new Google Sheet: sheets.google.com Set up your sheet with these columns: Column A: Prompt (your image generation prompts) Column B: Name (identifier for file naming) Example data: | Prompt | Name | |-------------------------------------------|-------------| | A serene mountain landscape at sunrise | mountain-01 | | Modern office space with natural lighting | office-02 | | Cozy coffee shop interior | cafe-03 | Copy the Sheet ID from the URL (the long string between /d/ and /edit) Step 4: Set Up Google Drive Folder Create a folder in Google Drive for your generated images Copy the Folder ID from the URL when viewing the folder Note: The workflow is configured to use a folder called "n8n workflows" Step 5: Import and Configure the Workflow Copy the provided workflow JSON In n8n, click Import from File or Import from Clipboard Paste the workflow JSON Configure each node as detailed below: Node Configuration Details: Start Workflow (Manual Trigger) No configuration needed Used to manually start the workflow Get Prompt from Google Sheet (Google Sheets) Document ID: Your Google Sheet ID (from Step 3) Sheet Name: Sheet1 (or your sheet name) Operation: Read Credentials: Select your "Google Sheets account" Double Output (Code Node) Purpose: Creates multiple variations of each prompt JavaScript Code: javascript const original = items[0].json; return [ { json: { ...original, run: 1 } }, { json: { ...original, run: 2 } }, ]; Customization: Add more runs for additional variations Loop (Split in Batches) Processes items in batches to manage API rate limits Options: Keep default settings Reset: false Create Image (HTTP Request) Method: POST URL: https://api.freepik.com/v1/ai/text-to-image Authentication: Generic → HTTP Header Auth Credentials: Select your "Header Auth account" Send Body: true Body Parameters: Name: prompt Value: ={{ $json.Prompt }} Split Responses (Split Out) Field to Split Out: data Purpose: Separates multiple images from API response Convert to File (Convert to File) Operation: toBinary Source Property: base64 Purpose: Converts base64 image data to file format Upload Image to Google Drive (Google Drive) Operation: Upload Name: =Image - {{ $('Get Prompt from Google Sheet').item.json.Name }} - {{ $('Double Output').item.json.run }} Drive ID: My Drive Folder ID: Your Google Drive folder ID (from Step 4) Credentials: Select your "Google Drive account" Step 6: Customize for Your Use Case Modify Duplicate Count: Edit the "Double Output" code to create more variations Update File Naming: Change the naming pattern in the Google Drive upload node Adjust Batch Size: Modify the Loop node settings for your API limits Add Image Parameters: Enhance the HTTP request with additional Freepik parameters (size, style, etc.) Step 7: Test the Workflow Ensure your Google Sheet has test data Click Execute Workflow on the manual trigger Monitor the execution flow Check that images are generated and uploaded to Google Drive Verify file names match your expected pattern Step 8: Production Deployment Set up error handling for API failures Configure appropriate batch sizes based on your Freepik API limits Add logging for successful uploads Consider webhook triggers for automated execution Set up monitoring for failed executions Freepik API Parameters Basic Parameters: prompt: Your text description (required) negative_prompt: What to avoid in the image guidance_scale: How closely to follow the prompt (1-20) numinferencesteps: Quality vs speed trade-off (20-100) seed: For reproducible results Example Enhanced Body: json { "prompt": "{{ $json.Prompt }}", "negative_prompt": "blurry, low quality", "guidance_scale": 7.5, "numinferencesteps": 50, "num_images": 1 } Workflow Flow Summary Start → Manual trigger initiates the workflow Read Sheet → Gets prompts and names from Google Sheets Duplicate → Creates multiple runs for variations Loop → Processes items in batches Generate → Freepik API creates images from prompts Split → Separates multiple images from response Convert → Transforms base64 to binary file format Upload → Saves images to Google Drive with organized names Complete → Returns to loop for next batch Contact Information Robert A Ynteractive For support, customization, or questions about this workflow: 📧 Email: rbreen@ynteractive.com 🌐 Website: https://ynteractive.com/ 💼 LinkedIn: https://www.linkedin.com/in/robert-breen-29429625/ Need help implementing this workflow or want custom automation solutions? Get in touch for professional n8n consulting and workflow development services.
By Robert BreenAI-powered cover letter generator with resume matching & Google Docs
This workflow generates a tailored cover letter using a provided resume and job description. Users submit a job description via form (or workflow input), the workflow uses an LLM to write a professional, casual cover letter, then creates and populates a Google Doc and redirects the user to download or review it. --- What You Must Update Before Running Resume Content Update the Configuration node to include your own resume text. This resume is injected directly into the prompt and used as the sole source of experience and qualifications. LLM Credentials The workflow uses OpenRouter with an OpenAI-compatible model. You must: Add your own OpenRouter API credentials Optionally change the model selection if desired Google Docs Credentials This workflow creates and edits Google Docs. You must: Connect your own Google Docs OAuth credentials Update the destination folder ID if you want files saved elsewhere --- What You Can Configure Prompt Tone & Constraints Edit the Write Cover Letter agent system prompt to adjust: Tone (more formal or more casual) Length Writing style constraints Execution Method The workflow supports: Manual execution via form submission Execution as a sub-workflow via workflow inputs
By Joel Gamble