U.S. Census Bureau housing-relevant APIs. Provides access to American Community Survey data, building permits, housing starts, homeownership rates, and available datasets.
Census MCP defines 5 tools with complete input schemas and descriptions. All tools follow verb_noun naming (census_*). Descriptions are substantive (100-200 chars), explaining what data is returned and common use cases. However, output schemas are not documented, LLMs cannot predict response structure. Parameters lack enums for constrained values (e.g., region, category_code). Error handling is minimal. The _apiKey parameter violates secret-injection pattern by exposing credentials as a tool parameter. No tool annotations (readOnlyHint, etc.) despite all being read-only operations.
Get American Community Survey (ACS) 5-year data from the U.S. Census Bureau. The core dataset for housing statistics including total housing units, median home value, owner-occupied units, median rent, and more. Common variable codes: B25001_001E (total housing units), B25077_001E (median home value), B25003_002E (owner-occupied), B25003_003E (renter-occupied), B25064_001E (median gross rent), B25071_001E (median rent as % of income).
List available Census Bureau datasets. No API key required. Useful for discovering dataset identifiers, descriptions, and available variables before querying specific data.
Get monthly building permits data from the Census Bureau residential construction survey. Tracks new privately-owned housing units authorized by building permits.
Get quarterly homeownership rates from the Census Bureau Housing Vacancy Survey (HVS). Reports the percentage of occupied housing units that are owner-occupied.
Get new residential construction data including housing starts, units under construction, and completions from the Census Bureau.
API key exposed as tool parameter (_apiKey). Credentials must use server-side secret injection via environment variables or vault. Agent traces log all parameters, secrets leak into logs and prompt history.
Output schemas not documented. LLMs cannot predict response structure (e.g., tableToObjects returns array of objects with dynamic keys). Without documented output, agents cannot plan downstream tool calls or extract required fields reliably.
Constrained parameters lack enums. 'region' (NE, MW, S, W), 'category_code' (TOTAL, 1UNIT), and 'time' format are free-form strings. LLMs will hallucinate invalid values. Use enum constraints or regex patterns.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 56 | 2026-07-28+ | v2 |
No tool annotations despite all tools being read-only. Add readOnlyHint: true to signal safety and enable agent optimizations. Current spec (2026-07-28) supports tool annotations.
Error handling is minimal. censusFetch() throws generic 'Census API error' without guidance on recovery. No distinction between retryable (rate limit, timeout) and user-fixable (invalid variable code) errors.