WhatsApp · Google Gemini 77 nodes
WhatsApp Master Agent
Multimodal WhatsApp agent: text, audio, image, and documents.
About this template
Full-featured WhatsApp agent that processes text, audio (transcription via Whisper), images, and documents. Uses Google Gemini as the primary LLM with OpenAI fallback, manages conversation context in Supabase, and can execute actions such as creating Google Calendar events and sending Gmail messages. With 77 nodes it covers media-type routing, human handoff, and automation pause by keyword.
What you get
- workflow.json (77 nodes) ready to import
- SQL schemas for conversation history and agent state
- README in PT-BR and EN
- ARCHITECTURE.md with media-type routing map
- Documented .env.example
- Commercial LICENSE
Prerequisites
- n8n self-hosted with internet access
- Configured Evolution API instance
- Google Gemini API key (Gemini 1.5 Pro recommended)
- OpenAI API key for Whisper (audio transcription)
- Supabase project with included schema
- Google account with Google Calendar and Gmail (OAuth2)
Built on
n8n Google Gemini OpenAI Whisper Evolution API Supabase Google Calendar Gmail