Skip to main content
Glama
chrischall

zillow-mcp

by chrischall
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

Bulk-fetch Zillow properties by zpid

zillow_bulk_get
Read-onlyIdempotent

Fetch up to 200 Zillow property records in one call using ZPIDs or homedetails URLs. Returns a per-ID success/error row, so one bad listing does not fail the batch.

Instructions

Fetch up to 200 Zillow property records in a single call — the "give me everything for these N saved homes" endpoint. Returns one structured row per input id (no pivoted side-by-side summary table — for 2-25 listings with a comparison summary use zillow_compare_properties). Each row is either { zpid, status: "ok", property } on success or { zpid, status, error_kind, retryable, error } on failure (status = error_kind) — one bad zpid never fails the whole call. Calls fan out concurrently against /homedetails/<zpid>_zpid/ (capped at 6 in flight, per issue #78, with retry-once-on-timeout per sub-request to absorb transient SW evictions). Big lists fan out bounded to 6 in flight and paced by a per-host requests-per-minute throttle (burst 20) so the batch doesn't trip Zillow's PerimeterX bot-wall (issue #90). If the bot-wall is hit, the blocked sub-requests are retried with exponential backoff; anything still blocked is reported with error_kind: "bot_challenge" (distinct from a missing listing) and the response carries a { blocked, retry_after_s } envelope so you can finish the rest in a second pass. The whole call is bounded by an overall hard deadline (issue #98): a single slow/hung row never wedges the server — when the deadline is reached any row that has not yet settled is returned with error_kind: "pending" and the response carries a { pending } count so you can re-run just those ids. The envelope also reports count / ok / errored.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
urlsNoZillow homedetails URLs/paths to fetch. 1..200.
zpidsNoZpids to fetch. 1..200. Provide either zpids or urls, not both.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv1.2.4
    • changedInput schema / properties / zpids / description
      Previous value: -"Zpids to fetch. 1..200. Provide either zpids or urls."New value: +"Zpids to fetch. 1..200. Provide either zpids or urls, not both."
    • changedInput schema / properties / zpids / items / anyOf
      Previous value: -[
-  {
-    "exclusiveMinimum": 0,
-    "maximum": 9007199254740991,
-    "type": "integer"
-  },
-  {
-    "type": "string"
-  }
-]New value: +[
+  {
+    "exclusiveMinimum": 0,
+    "maximum": 9007199254740991,
+    "type": "integer"
+  },
+  {
+    "pattern": "^\\d+$",
+    "type": "string"
+  }
+]
  2. Changed1 schema field changedv1.0.0
    • changedInput schema / $schema
      Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
  3. First observedv0.11.1

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/openWorld, and the description layers on substantial extra behavior: 6-in-flight concurrency cap, retry-once-on-timeout, per-host RPM throttle with burst 20, PerimeterX bot-wall retries with exponential backoff, a hard overall deadline, and per-row error isolation. It also names the failure taxonomy (bot_challenge vs pending vs missing) and the blocked/retry_after_s and pending envelopes.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded correctly: purpose, then the sibling alternative, then return shape, then internals. It is dense and somewhat over-stuffed, and the 6-in-flight cap is stated twice ('capped at 6 in flight' then 'fan out bounded to 6 in flight'), plus issue-number citations (#78/#90/#98) are marginal for an agent, but nearly every sentence carries real behavioral information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema, the description fully carries the return-value burden: it specifies the two row shapes, the status/error_kind/retryable fields, and the count/ok/errored/blocked/pending envelope. Combined with annotations covering safety, an agent has everything needed to call and interpret the tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so both urls and zpids are already documented (including maxItems/minItems and mutual exclusivity), which sets the baseline at 3. The description adds meaning beyond the schema by mapping inputs to outputs one-to-one ('one structured row per input id') and clarifying that a single bad id does not fail the batch.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb, resource, and hard scope cap ('Fetch up to 200 Zillow property records in a single call') and explicitly distinguishes itself from the sibling zillow_compare_properties. An agent can route correctly without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives an explicit when-not rule and names the alternative with its own selection condition ('no pivoted side-by-side summary table — for 2-25 listings with a comparison summary use zillow_compare_properties'). It also frames the batch scenario ('all saved homes') that selects this tool, leaving nothing to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools