US-Weather-Hazards
Top drivers
⌁ mcp.call("adw-252") vADW-252-live-1.0 How heavy is the current US severe-weather alert load?
US-Weather-Hazards
Top drivers
⌁ mcp.call("adw-252") vADW-252-live-1.0 A logistics dispatch agent consumes ADW-252 hourly and automatically re-routes long-haul truck dispatches away from corridors covered by Extreme or Severe NWS alerts (weight ≥3 in the severity-weighted score) when hazard_load_score exceeds 60, logging the methodology_version so the routing decision is traceable in the fleet management audit trail. The backtest shows hazard load has ranged from 18.8 to 41.8, with a mean of 30.3 — a current reading of 18.8 signals below-average load, while a spike to 60+ would represent a statistically elevated event warranting automated re-routing. Source_lineage (NWS api.weather.gov active alerts) confirms the agent is acting on the same authoritative feed FEMA uses, not a third-party aggregation.
A regional VP of operations at a grocery distribution network uses the extreme_alerts and severe_alerts counts from ADW-252 to decide whether to pre-position perishable buffer stock at distribution centers in the alert footprint before a major weather event, a decision that previously depended on individual store managers monitoring local weather apps inconsistently. The severity-weighted composite converts the NWS raw alert volume into a single prioritization score, allowing the VP to act on a documented threshold rather than subjective judgment — improving both the consistency and the defensibility of pre-storm inventory decisions.
Severity-weighted (Extreme4/Severe3/Moderate2/Minor1) active-alert sum, scaled
Version ADW-252-live-1.0 · validated to beat a naive baseline · benchmark: Weather-channel feeds (consumer); NWS is the authoritative open source
One call returns the answer with its reasoning attached — the live Intelligence Object for ADW-252.
{
"product_id": "ADW-252",
"entity": "US-Weather-Hazards",
"score": 29.5,
"trend": "stable",
"confidence": 0.65,
"top_drivers": [
{
"factor": "active_alerts",
"contribution": 198
},
{
"factor": "extreme_alerts",
"contribution": 1
},
{
"factor": "severe_alerts",
"contribution": 65
}
],
"methodology_version": "ADW-252-live-1.0",
"freshness": "2026-06-26T22:00:12.731Z",
"coverage": "US active NWS weather alerts, severity-weighted",
"source_lineage": [
"api.weather.gov/alerts/active (NWS, keyless)"
],
"allowed_use": "evaluation, commercial",
"validation_status": "descriptive"
} Every product conforms to the Intelligence Object Model — typed, versioned, and discoverable.
Dashboard
Read the score + drivers in the console.
REST API
/v1/intelligence/adw-252
MCP tool
adw.adw_252
Marketplace
Discoverable by any MCP agent via the MCP registry.
White-label
Embed under your own brand (Platinum).
Enable organizations to prioritize and allocate capital expenditure for climate resilience projects based on quantified physical risk exposure.
Method: Aggregate geospatial hazard data (flood, fire, heat) with asset location and value inputs to calculate a risk-adjusted return on investment score for each potential adaptation intervention.
Are US climate-disaster costs trending above historical norms?
Method: z-score of latest annual cost vs full 1980-present history → clamp(50 + z×15, 0, 100)
Is river-discharge building toward flood risk on the US Gulf Coast?
Method: Peak-vs-current 7d river-discharge ratio, scaled