HVAC WhatsApp AI Reception n8n Workflow (Gemini, Claude, API)

Turn your WhatsApp inbox into an AI HVAC receptionist—automatically

This HVAC WhatsApp AI Reception n8n Workflow answers incoming WhatsApp messages with an AI concierge. It receives messages (text, voice notes, and photos) via WhatsApp Cloud API, transcribes and describes content using OpenRouter (Gemini/Claude), then drafts smart replies using Anthropic Claude with PostgreSQL chat memory—sending responses back as up to three WhatsApp messages.

What this workflow does

• Triggers on every new WhatsApp message received through your Meta webhook.
• Routes by message type: passes through text as-is, transcribes voice notes via OpenRouter (Gemini), and describes photos via OpenRouter (Claude) while preserving any image caption.
• Normalizes and stores input in an n8n Data Table named “whatsapp inbox”, saving the business phone number and the resulting text.
• Waits briefly for multiple rapid messages, then checks whether the current run is the newest one for that phone number.
• Merges queued lines for that customer, clears them from the Data Table, marks the WhatsApp message as read, and shows a typing indicator.
• Generates a reply using a LangChain agent powered by Anthropic Claude, with PostgreSQL chat memory (and optional SerpAPI web search and calculator tools).
• Splits the response into up to three short WhatsApp messages and sends them back via WhatsApp Cloud API.

Use cases

• Answer HVAC service requests instantly: “Are you available today?” “Pricing for maintenance?”
• Handle voice-note troubleshooting details (e.g., “unit is cooling but not heating”).
• Process photo-based diagnostics (e.g., thermostat display, system labels) and respond with next steps.

Technical details

• WhatsApp Cloud API (Facebook Graph API header auth) + Meta webhook trigger
• n8n workflow nodes: set, code, wait, filter, switch, WhatsApp
• OpenRouter for transcription (Gemini) and image description (Claude)
• LangChain agent with Anthropic Claude
• PostgreSQL chat memory for conversational context