n8n WhatsApp Reply Bot: WaAPI + GPT-4o-mini Auto-Responses

Instant WhatsApp auto-replies with WaAPI + GPT-4o-mini (including “ping → pong”)

Turn inbound WhatsApp messages into helpful, concise responses automatically. This n8n workflow listens for WaAPI message events, sends “seen” and typing indicators, and then uses an OpenAI gpt-4o-mini agent with short-term conversation memory to generate a reply per chat—plus a built-in ping/pong health check.

What this workflow does

• Receives a POST webhook from WaAPI and continues only when the incoming event type is a message.
• Waits briefly, then uses WaAPI to mark the chat as seen.
• Waits again, then uses WaAPI to display a typing indicator while the response is generated.
• Checks the incoming message text: if it equals “ping”, it immediately sends “pong” back to the same chat via WaAPI.
• For all other messages, forwards the message to an OpenAI chat model (gpt-4o-mini) to create a short, language-matching reply using buffer memory keyed by chat ID (so conversations stay coherent during the short term).
• Stops typing and sends the generated response back to the WhatsApp chat using WaAPI.

Use cases

• Customer support triage: respond quickly to common WhatsApp inquiries while keeping a short conversation context.
• Lead engagement: acknowledge messages with typing/seen indicators and provide concise answers automatically.
• Operational monitoring: use the built-in ping → pong to verify the workflow and WaAPI webhook connectivity.

Technical details

• n8n nodes/tools: webhook, if, wait, waapi (WaAPI send/typing/seen), and n8nn8n-nodes-langchainagent for the OpenAI chat agent.
• OpenAI integration: OpenAI chat model (gpt-4o-mini) with buffer memory keyed by chat ID.

Setup checklist

• Connect your WaAPI credentials in n8n and provide a valid instanceId.
• Add an OpenAI API credential for gpt-4o-mini.
• Copy the workflow webhook URL (path /waapi-chatbot) into your WaAPI webhook configuration so WhatsApp message events are delivered to n8n.