# hazards_rank_geographies Rank counties or tracts by a composite or hazard-specific NRI metric. | Field | Value | | --- | --- | | Category | Hazards (`hazards`) | | Exposure | public | | MCP tool | `hazards_rank_geographies` | | Documentation | https://www.agentdatasets.com/docs/tools/hazards_rank_geographies | ## Description Rank counties or tracts by a composite or hazard-specific NRI metric. The service selects the newest release held for each geography, excludes missing metric values, and orders descending with a stable FIPS tie-break. Each row carries release vintage/date and `as_of`; `data.metric_unit` distinguishes scores from USD/year expected losses. Keep level, metric, and state fixed when paging with `pagination.next_cursor`. ## Parameters Input schema advertised to MCP clients and mirrored by the REST query string. | Parameter | In | Type | Required | Default | Constraints | Description | | --- | --- | --- | --- | --- | --- | --- | | level | input | `"county" \| "tract"` | yes | — | — | Geography level to rank: county or tract. | | metric | input | `string` | yes | — | — | Metric: risk_score, eal_total, sovi_score, resl_score, or a FEMA hazard code/name plus _risk_score or _eal_total, such as wildfire_eal_total. | | state | input | `string \| null` | no | null | — | Optional state name, postal abbreviation, 2-digit FIPS, or state_ geo_key resolved through the shared geography layer. | | limit | input | `integer` | no | 10 | min 1, max 100 | Ranked geographies to return (1–100). | | cursor | input | `string \| null` | no | null | — | Opaque cursor from the previous pagination.next_cursor. | ## Provenance - `fema-nri` — Federal Emergency Management Agency, National Risk Index county and census-tract profiles with composite risk, expected annual loss, social vulnerability, community resilience, and 18 hazard-specific scores and ratings (U.S. Government work, public domain (17 U.S.C. § 105)). See https://www.agentdatasets.com/docs/attribution.md Dataset registry: `hazards_nri_profiles`.