Hello Everyone,
I want to create one agent that consults files that are hosted in a google drive and after that give one resume or answer questions about the content in the files, I have docx and csv files on google drive.
I have made some texts with Google Drive MCP, but it works just with google files.
Does anyone know if it is the best way to do that? or maybe I can try another type of logical on my workflow.
Hi @Rodrigo_Americo Welcome to the community!
I have done it and it is pretty awesome , create a folder in your google drive and add a trigger in your workflow that once there is some file added into that workflow the flow would start, and for now just download that file and classify it based on .txt .pdf and extract contents from it and either you can create a vector database for it or you can just use data table to add all the extracted data, and once it is done your AI chatbot would be able to fetch data from that data table and would assist users with the help of the data as it would be used as its context. Let me know if you have any specific problem in that
Yeah the MCP only works with native Google files so you’ll need to download them with the Google Drive node instead. CSV is easy, just pipe it through Extract from File, for .docx you’ll either need the community node n8n-docx-converter or convert them to Google Docs format first. Once you have the text just feed it into your agent directly or build a quick RAG pipeline with a vector store if there’s a lot of content.