Answer WhatsApp customer questions and escalate to humans

Case study takeaway: an n8n workflow takes each incoming WhatsApp message and assembles the conversation history plus the company knowledge base. OpenAI sketches a reply grounded in that content, and a confidence check decides whether it can be answered reliably. The README gives no measured results, only design goals: cover routine questions, give round-the-clock coverage, and keep a human on unreliable cases.

Agent or tool: n8n
Topic: Research & Data
Reported on: GitHub
Added: 2026-10-15

Why it matters

Research is mostly gathering, cleaning and comparing. Handing that groundwork to an agent leaves the person with the part that needs judgment: deciding what the findings mean. It is one of 201 sightings filed under Research & Data on Agent Sightings (number 180 by interest), and one of 37 involving n8n.

How you could try it

Here is a reusable prompt, written by us in our own words, that captures the task. Paste it into n8n or any similar agent and adjust the details to your situation.

Research brief: answer WhatsApp customer questions and escalate to humans. Prefer primary sources, include links and confidence notes, and do not contact anyone.

Saved

Tip: Start small: run it once on a low-stakes example, read what the agent did, and only then let it run on real work.

Source

Read the original first-hand account on GitHub: https://github.com/zezo640/whatsapp-ai-support-agent-n8n. Our summary is a paraphrase; details are in the source.

Built an AI agent? Feature your agent on Agent Sightings, free submission or a featured spot.

Browse more like this

More in Research & Data · more with n8n · all browse pages · industry guides

Related sightings

Turn a business website into a WhatsApp support agent · n8n · Research & Data
One practitioner notes that an n8n workflow accepts any business URL, scrapes the full site via AI to create a knowledge base, afterward powers a…
50 AI agents in Minecraft built a government, a ration ledger and a no-blame inquiry about a chicken · GPT-6.1 Sol agents · Research & Data
YC Labs, working with Olam Labs, put 50 GPT-6.1 Sol agents in a survival Minecraft world for three hours with the goal 'Build a thriving…
Pensive offers personalized AI teaching assistants for higher ed with Claude · Claude · Research & Data
Pensive — Read the case study to see how they build on the Claude Platform. Pensive aims to build an AI Teaching Assistant (AI TA) that supports a…
From weeks to a day: Everstar on Microsoft Foundry with Claude · Claude · Research & Data
Everstar Foundry Models — Everstar built Gordian on Claude in Foundry Models, cutting nuclear permitting timelines 80–96% and completing 4–6 weeks…
Triage car repair requests and reply to customers · n8n · Shopping
A published walkthrough shows an n8n workflow receives car repair requests from a web form. The AI classifies each by job type, urgency, and…
Invoices filed the moment they arrive · n8n · Finance
An automation reads invoice PDFs or photos from email, extracts key fields, files them in Drive with consistent names and logs them in a sheet.

← All 2754 use cases