Turn long PDFs into clear image insights. The flow pulls a PDF from cloud storage, extracts every image, sends each image to AI for review, and compiles all findings into one text file. Ideal for marketing teams, operations staff, and QA checks that need fast visual summaries.

Under the hood, a manual start fetches a PDF from Google Drive. An HTTP call to ConvertAPI extracts images as JPG files. A split step creates one item for each image. Each image URL is fed to OpenAI GPT 4o with a prompt for detailed analysis. The output pairs each image link with the AI summary, then a code step merges all results. A final node generates a single txt file ready for download or sharing.

Have active accounts for Google Drive, OpenAI, and ConvertAPI. Expect faster content reviews, cleaner notes, and less context switching. Use it for design reviews, catalog checks, marketing proofing, and training materials. Plan for retries if the file is large or the image service is busy, and adjust the prompt to match your brand or task.

• Manual start that you can swap with another trigger later
• Download a selected PDF from Google Drive using the file ID
• Extract all images from the PDF via ConvertAPI and return JPG files
• Split the Files array so each image is handled separately
• Map each image URL for clean use in the AI step
• Analyze images with OpenAI GPT 4o using a customizable prompt
• Combine each image link with its AI notes into clear text
• Merge all summaries into one document with a code step
• Create a final txt file ready for download or sharing

• Reduce manual review from 2 hours to 10 minutes for a typical PDF with many images
• Automate up to 90 percent of repetitive image notes and summaries
• Improve consistency of visual reviews by standardizing the AI prompt
• Connect Google Drive, ConvertAPI, and OpenAI in one flow
• Scale to long documents with dozens of images without extra effort

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Google Drive, OpenAI and ConvertAPI. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager or by double clicking the Google Drive node named Get pdf file, click Create new credential, then follow the on screen steps to connect your Google Drive account. Confirm access to the folder that holds your PDFs.
4. Open the Get pdf file node and set the file to download. Use the file picker if available or paste the file ID. Save the node and run a quick test to confirm the PDF downloads.
5. Double click the HTTP Request node named Extract pdf image. In Credential to connect with choose Create new credential and pick HTTP Header Auth. Add the header required by ConvertAPI for your secret key as described in your ConvertAPI account, then save.
6. In your ConvertAPI account, copy your API key or secret. If ConvertAPI uses a specific header name for the key in your plan, enter that exact name and value in the credential. Save the credential and test the node with a small PDF.
7. Open the Analyze image node. In Credential to connect with choose Create new credential for OpenAI. Paste your API key from the OpenAI dashboard and save. Verify the model is set to GPT 4o.
8. Review the prompt text in Analyze image. Update the wording so it fits your use case, such as brand checks, layout issues, or product detail review. Keep instructions simple and direct.
9. Run the workflow with the manual trigger. After the HTTP step, check the Split node output to see one item per image. Confirm image URLs are present.
10. Open the Code node and preview the merged content output. Ensure each section includes the image link followed by the AI summary.
11. At the end, open the Output content to a .txt file node and download the file. Review for clarity. If ConvertAPI returns 503 errors, wait a few seconds and run again or add retry logic in the HTTP node.
12. Optional swap the manual start with a Google Drive trigger if you want the analysis to run when a PDF is added to a folder.