Zillow Leads Property Data

io.github.kmatataPublished by GitHub account kmatatav1.0.8Updated Oct 24, 2026

MCP server for the zillow-leads-property-data actor: Zillow leads with agent contacts and history.

Overview

AI-generated overview

Local MCP server that runs the Zillow Leads & Property Data Apify actor, returning Zillow listings with agent contacts, price and tax history, and distress…

What you can do
  • Run the Zillow leads actor in catalog mode to get a cached, already-enriched snapshot for a metro without waiting.
  • Order live collection in custom_search or recent_activity modes and retrieve the resulting listing rows.
  • Get bare listings fields (address, price, beds/baths, status, area) or enriched rows with agent/broker contacts, price and tax history, foreclosure signals, schools, HOA and resoFacts.
  • Carry zpids and MLS IDs forward between orders to avoid paying twice for the same row.
When to use it
Useful for real-estate lead generation and property research where an assistant should pull Zillow listing data with agent contact details and long price/tax histories, either from a cached metro snapshot or a live collection order.
Requirements
Runs as a local stdio process on the user's machine (desktop only). It drives an Apify actor, so an Apify account and Apify API credentials are needed; the actor is billed per event. Network access to Apify and Zillow is required.
Before you install
The actor is pay-per-event: $0.70 per 1,000 bare rows and $1.20 per 1,000 enriched rows, so live orders spend money on the user's Apify account. Data is scraped from Zillow through a real browser session, and enriched rows include agent names, phone numbers and emails, which are personal contact data. The repository itself is documentation and samples only and does not run the actor.

Installation

In SourceWeft

  1. Open Zillow Leads Property Data in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

zillow-leads-property-data

Code samples, a sample dataset, and a full workflow notebook for the Zillow Leads & Property Data Apify actor.

The actor pulls Zillow listings enriched with agent/broker contact info, full price-history timelines, 20yr+ tax history, foreclosure/distress flags, schools, and the resoFacts long tail (heating/cooling/construction). Pay-per-event pricing, no subscription: $0.70 per 1,000 bare rows, $1.20 per 1,000 enriched rows.

Try it in 30 seconds, no signup beyond an Apify account

The actor's catalog mode returns a cached, already-enriched snapshot instantly, no bounding box, no wait:

json
{ "mode": "catalog", "metro": "Chicago" }

Run it from the Apify Console or via apify-client (see examples/ below).

What's in this repo

PathWhat it is
sample/sample_listings.csv20 real enriched rows (Chicago/Houston/Phoenix) — flat, spreadsheet-ready
sample/sample_listings.jsonlSame rows, one JSON object per line, with a nested price_history array per listing
examples/python_quickstart.pyCall the actor from Python via apify-client, wait for results, print a summary
examples/node_quickstart.jsSame, in Node, via apify-client
mcp-server/One-tool MCP server for Claude Desktop, Cursor and any MCP client; see mcp-server/README.md
notebook/zillow_leads_workflow.ipynbA realistic lead-gen workflow: run the actor, dedupe against a prior export, filter to rows with agent phone numbers, export a clean CSV

Field reference

Two depths, listings (bare) and enriched (the default):

  • listings: address, price, beds/baths, status, lot/living area, listing-type flags — whatever Zillow's own search API returns, no detail-page fetch involved. See sample_listings.csv for the shape (a subset of the columns shown there — the sample above is enriched-depth, listings-depth rows omit everything past the bare fields).
  • enriched (default): everything in listings, plus agent/broker contact (name, phone, email, brokerage, MLS attribution), full price-history and 20yr+ tax-history timelines, foreclosure/distress signals, assigned + nearby schools, HOA fees, and the resoFacts long tail (heating/cooling/parking/construction).

Full field list, pricing table, and every input mode (catalog / custom_search / recent_activity) are documented on the actor's own README.

Dedup across repeat orders

Every delivery writes a DEDUP_UPDATE record to the run's key-value store: the union of the zpids and MLS IDs you already had plus everything just shipped. Feed those two arrays back into your next order's dedupZpids/dedupMlsIds fields and you're never charged for the same row twice. See notebook/zillow_leads_workflow.ipynb for a worked example.

Use it from an AI agent (MCP)

This actor is callable as a tool via Apify's hosted MCP server (https://mcp.apify.com) or the local @apify/actors-mcp-server, no manual input-form filling required — catalog mode (the default) is a good agent smoke test since it returns cached rows in seconds with no polling loop needed. See the actor's own README for the full agent-facing input schema and the fire/poll/fetch pattern for a live-collection order.

Questions / issues

This repo is documentation and samples only — it doesn't run the actor itself. For a specific order that isn't behaving as expected, check that run's key-value store STATUS record first (the actor writes a plain-English reason there for any zero-row or partial result), then open an issue here or reach out via the actor's Apify Store page.

Source: README.md at commit 01fa450

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.8LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts