1 ready-to-build workflow

AI agent workflows for Real estate

Qualify portal leads and match inventory before a rival even calls back.

Each recipe below is expressed only in VegaDūta's real workflow building blocks — Trigger, Agent, Knowledge, Condition, Approval, Output, Tool (MCP), Voice, Device, Loop, Code and Parallel — so it maps 1:1 to something you can assemble in the Workflow designer. Consequential actions always pass a human Approval step.

1. Portal lead → qualify → matched properties → site visit

Leads from property portals go cold while agents are on viewings.

Trigger: Webhook (portal lead)

How the workflow runs

  1. Trigger

    Portal sends a lead. A Webhook trigger receives the enquiry (budget, area, BHK).

  2. Agent

    Qualify the requirement. An Agent node clarifies budget/location/timeline over WhatsApp.

  3. Knowledge

    Match live inventory. A Knowledge node searches your current listings for genuine matches.

  4. Output

    Send matches + book a visit. An Output node shares 2–3 properties and offers site-visit slots.

  5. Approval

    Hand a hot lead to the agent. An Approval/Output node alerts the right agent when the lead is sales-ready.

Channels & connectors

  • Webhook
  • WhatsApp
  • Knowledge base
  • Approval / HITL

Outcome

Every portal lead is engaged and pre-qualified in minutes; agents only spend time on ready buyers.

Why it helps

Response speed and relevant matches are what convert portal leads — this does both before the lead cools.

Build spec

1 agent1 qualification agent; the inventory match is Knowledge and the hand-off is an Approval node.

System prompt (paste-ready)

You are a real-estate lead-qualification agent. Given a portal enquiry {name, budget, area, bhk, timeline}, hold a short WhatsApp conversation to confirm budget, location, configuration and buying timeline. Then use Knowledge to find 2–3 genuine matches from current listings — never present a property outside the stated budget/area, and never invent inventory. Score the lead's readiness as hot/warm/cold and offer site-visit slots. Hand hot, sales-ready leads to a human agent.

MCP connectors

  • WhatsApp — conversation channel
  • Property portal — inbound via Webhook trigger

Built-in tools

  • knowledge_search (current live listings)

Guardrails

  • Only real, matching inventory is shown — no invented or off-budget listings
  • A hot lead is handed to the right human agent via the Approval node before any commitment
  • Budget and area constraints from the enquiry are respected in every match

Output & delivery

The agent shares 2–3 Knowledge-matched properties on WhatsApp with site-visit slots, and routes sales-ready leads to the assigned agent through the Approval node.

Related industries

Build one of these in minutes

The sandbox gives you a live agent workspace — no account, no card. Or head back to the full catalogue and compare patterns across every industry.

See how VegaDūta compares to n8n, Dify and BotpressWhy VegaDūta's architectureEstimate your WhatsApp API costs