What This Automation Does
This workflow takes images and PDFs as input and sends them to Google Gemini AI for analysis.
The main problem it solves is saving time and effort when reviewing many marketing assets.
The output is detailed content descriptions and key facts extracted from each file.
This helps users get fast, consistent insights without manual work.
How This Workflow Works (Inputs → Processing → Output)
Inputs
- Image URLs with prompts explaining the requested analysis.
- PDF file URLs for content extraction.
Processing Steps
- Split multiple input items into singles for processing.
- Filter only items marked to process.
- Download images or PDFs using HTTP nodes.
- Convert binary data to base64 encoding.
- Send base64 data with prompts to Google Gemini API.
- Receive AI-generated text with insights.
Output
- AI descriptions and summaries of image or PDF content.
- Useable text for marketing teams to quickly understand asset details.
Who Should Use This Workflow
Anyone who manages many images or PDFs needing quick content review.
Good for content managers, marketers, or teams with fast delivery needs.
It helps reduce hours spent on manual analysis and speeds campaigns.
Tools and Services Used
- Google Gemini (PaLM) API: For AI content understanding.
- Unsplash API or direct image URLs: For sourcing images.
- n8n Automation Platform: To build and run the workflow.
- HTTP Request nodes: To access APIs and download files.
Beginner Step-by-Step: How to Use This Workflow in n8n
Importing and Setting Up the Workflow
- Download the workflow file using the Download button on this page.
- Open the n8n editor and use “Import from File” to add the workflow.
- In the imported workflow, add your Google Gemini API Key in the HTTP Request node credentials.
- If URLs, prompt texts, or other settings are present in Set nodes, update them as needed for your media.
Testing and Activating
- Run the workflow manually once using the Manual Trigger node.
- Check the outputs to see if image and PDF data processed successfully.
- Fix any credential or URL issues if errors happen.
- Finally, activate the workflow in n8n to run automatically or on demand.
You now have a running media analyzer using Google Gemini AI without building it yourself.
Consider using self-host n8n for full control over data and uptime.
Common Workflow Inputs and Outputs
- Input: Array of objects with URL and prompt fields for images.
- Input: PDF file URLs for direct analysis.
- Output: Text responses with descriptions or color info for images.
- Output: Summaries of text content extracted from PDFs.
Potential Issues and How to Fix Them
- Authentication errors come from wrong or missing API keys; recheck credentials in HTTP nodes.
- Download failures usually mean bad URLs or access limits; confirm URLs in Set node are correct and public.
- Base64 errors happen if binary data isn’t properly converted; verify Extract From File nodes use “binaryToProperty” operation.
- AI response errors can occur if prompts are empty or not forwarded correctly; examine prompt fields and test with sample text.
Customization Ideas
- Edit prompts in the Set node to ask the AI specific questions about images or documents.
- Add your own PDF URLs in the HTTP Request node for document input.
- Adjust batch sizes or splitting logic to manage large input volumes efficiently.
- Switch Google Gemini models by changing model names in HTTP requests.
- Add error checks after downloads to skip broken images or files.
Summary
✓ Saves hours by automating image and PDF content review using Google Gemini AI.
✓ Processes single or multiple files with tailored prompts.
→ Outputs easy-to-understand text insights for marketing use.
→ Simplifies managing media with n8n’s automation platform.