What it does
An n8n template with two workflows. The first imports product documentation from Google Docs, splits it into chunks, creates OpenAI embeddings and stores them in a MongoDB Atlas vector store. The second answers questions by searching that store and replying with a GPT-4o-mini agent that keeps short-term memory.
Use cases
- 01Answer agent questions from product manuals
- 02Index Google Docs into a vector store
- 03Give support staff a chat over internal docs
Connects
Google Docs · AI Agent · Embeddings OpenAI · OpenAI Chat Model · Simple Memory · Recursive Character Text Splitter · Default Data Loader · MongoDB Atlas Vector Store