AI-powered auto-generate exam questions and answers from Google Docs with Gemini
Needs changesWhat this workflow does
This AI workflow has 32 nodes and runs manually, when you click “Execute workflow”. Along the way it connects Qdrant Vector Store, Embeddings OpenAI, Google Docs and Google Gemini Chat Model and 2 more. It also calls qdrant_url over HTTP.
How it runs
Trigger: Manually, when you click “Execute workflow”.
- Refresh collection (HTTP Request): calls qdrant_url (POST)
- Get Doc (Google Docs): get
- Converto di MD (Code): runs custom JavaScript code
- Closed questions (Basic LLM Chain): sends a prompt to an LLM, with model: Google Gemini Chat Model; output parser: Item List Output Parser
- Convert to File: converts data to a file (to text)
- Open questions (Basic LLM Chain): sends a prompt to an LLM, with model: Google Gemini Chat Model; output parser: Item List Output Parser
- Loop Over Items (Loop Over Items (Split in Batches)): loops over the items in batches
- Qdrant Vector Store: stores documents in the Qdrant vector store, with embeddings: Embeddings OpenAI; document loader: Default Data Loader
- Loop Over Items (Loop Over Items (Split in Batches)): loops over the items in batches
- Answer and create options (AI Agent): runs an AI agent, with model: Google Gemini Chat Model; 1 tool (RAG); output parser: Structured Output Parser
…plus 4 more steps. See the full node list below.
Services it uses
HTTP calls to: qdrant_url
Before you upgrade to n8n 3.0
1 thing in this workflow is removed in n8n 3.0. Fix it before you upgrade:
- AI Agent v1 (old agent modes) in Answer and create options (v1.9): replace with Latest AI Agent node (Tools Agent). For SQL Agent, use a Postgres or MySQL tool with a recent AI Agent.
Rather not rewrite it? Many of these flows can be rebuilt in Make (free plan).
| Removed in 3.0 | Replace with |
|---|---|
| AI Agent v1 (old agent modes) Answer and create options (v1.9) | Latest AI Agent node (Tools Agent). For SQL Agent, use a Postgres or MySQL tool with a recent AI Agent. |
Nodes used (19)
- Manual Trigger
n8n-nodes-base.manualTrigger - Qdrant Vector Store
@n8n/n8n-nodes-langchain.vectorStoreQdrant - HTTP Request
n8n-nodes-base.httpRequest - Embeddings OpenAI
@n8n/n8n-nodes-langchain.embeddingsOpenAi - Default Data Loader
@n8n/n8n-nodes-langchain.documentDefaultDataLoader - Token Splitter
@n8n/n8n-nodes-langchain.textSplitterTokenSplitter - Code
n8n-nodes-base.code - Google Docs
n8n-nodes-base.googleDocs - Vector Store Retriever
@n8n/n8n-nodes-langchain.retrieverVectorStore - Convert to File
n8n-nodes-base.convertToFile - Google Gemini Chat Model
@n8n/n8n-nodes-langchain.lmChatGoogleGemini - Item List Output Parser
@n8n/n8n-nodes-langchain.outputParserItemList - Loop Over Items (Split in Batches)
n8n-nodes-base.splitInBatches - Structured Output Parser
@n8n/n8n-nodes-langchain.outputParserStructured - Vector Store Question Answer Tool
@n8n/n8n-nodes-langchain.toolVectorStore - Basic LLM Chain
@n8n/n8n-nodes-langchain.chainLlm - Question and Answer Chain
@n8n/n8n-nodes-langchain.chainRetrievalQa - AI Agent
@n8n/n8n-nodes-langchain.agent - Google Sheets
n8n-nodes-base.googleSheets
From the author
This workflow automates the creation of exam questions (both open-ended and multiple-choice) from educational content stored in Google Docs, using AI-powered analysis and vector database retrieval This workflow saves educators hours of manual work while…