Automation library · Chat assistants

Build a multimodal WhatsApp assistant with OpenAI and Gemini

Started by an incoming WhatsApp message, and works across WhatsApp. It puts OpenAI behind a retrieval step, so answers are drawn from documents you supply rather than from the model alone. 23 steps, 1 branch: large, and worth reading before you trust it with anything live.

LargeCalls a model: OpenAIRetrieval

23 stepsLarge: read it before trusting it with anything live
1Branch: places the flow splits
0Loops or waits
1Service it connects to

The flow

Starts it 1Model 7Logic 7Service 8

The steps in the order they run, from what starts it.

  1. 1WhatsApp TriggerWhats App triggerStarts it
  2. 2OpenAI Vision ModelLm Chat Open AiModel
  3. 3OpenAI Chat ModelLm Chat Open AiModel
  4. 4Conversation MemoryMemory Buffer WindowModel
  5. 5WikipediaTool WikipediaModel
  6. 6Split out messagesSplit OutLogic
  7. 7Analyze image with GPTChain LlmModel
  8. 8AI AgentAgentModel
  9. 9ConfigurationSetLogic
  10. 10Normalize user messageSetLogic
  11. 11Format reply for WhatsAppSetLogic
  12. 12Route by message typeSwitchLogic
  13. 13Send WhatsApp replyWhats AppService
  14. 14Get audio URLWhats AppService
  15. 15Get video URLWhats AppService
  16. 16Get image URLWhats AppService
  17. 17Get text messageSetLogic
  18. 18Download audioHttp RequestService
  19. 19Download videoHttp RequestService
  20. 20Download imageHttp RequestService
  21. 21Transcribe audio with OpenAIOpen AiModel
  22. 22Describe video with GeminiHttp RequestService
  23. 23Extract video descriptionSetLogic

How it's built

Starts on
An incoming WhatsApp message
Steps
23, with 1 branch
Size
Large: read it before trusting it with anything live
Connects to
WhatsApp
Model
OpenAI
Retrieval
Answers are drawn from documents you supply, not from the model alone

Questions

What does it need to run?

An n8n account, plus your own WhatsApp account, and a key for OpenAI. You connect your own accounts after importing it; none are included.

What starts it?

It starts on an incoming WhatsApp message.

Does it use AI?

Yes. It calls OpenAI. Its answers draw on documents you give it, not on the model alone. Model calls are billed by the provider, not by n8n.

How long does it take to set up?

Take your time. At 23 steps it is large: read it before trusting it with anything live.

How do I import it?

Copy the workflow, open a new workflow in n8n and paste. The steps appear on the canvas, ready for your own accounts.

Before you run it

Nothing here has been tested by us. It is somebody's working automation, read and described from its own steps. Treat the connections it wants as a list of accounts it will touch, and read the 23 steps before you point it at anything that matters.