Google Drive, Pinecone, OpenAI, RAG?

[Jamie_C]

I have this working:

My Goal: Embed ALL files in a Google Drive folder to Pinecone. I can make this work one at a time, but I would like to simply have it look at what is in the folder and parse/embed all the pdfs and docs.

Notes: The embedding from Google Drive does not need to be monitored. I am perfectly fine with manually kicking off that embedding process when I add new files.

I want this to be a research assist of sorts. I add articles related to the topic as well as papers I write on the topic. This agent then helps me find references in the articles and revise the specific parts of my paper(s) as requested.

Thoughts? Is this easy? Hard? somewhere in between? Did I miss that someone has already done this and will feel silly when one of you points me to an JSON file to import the workflow? lol.

Thanks!

[n8n]

It looks like your topic is missing some important information. Could you provide the following if applicable.

• n8n version:
• Database (default: SQLite):
• n8n EXECUTIONS_PROCESS setting (default: own, main):
• Running n8n via (Docker, npm, n8n cloud, desktop app):
• Operating system:

[ihortom]

Hey @Jamie_C , your issues sedems to be narrowed down purely to being able to download PDFs from Google Drive. Here’s how you can do it.

Beware though that if you have many and large files you might want to introduce loop and/or sub-workflow to prevent your n8n instance crashing.