Next 24 hours
Proof24-hour local hazard closure
What is the overall weather risk for my region over the next 24 hours?
Your weather & hazard view
Weather Lab
See live source posture, short-term risk checks, medium-term hazard watches, and long-term climate/disaster signal hypotheses.
Your weather & hazard view
Weather Lab
See live source posture, short-term risk checks, medium-term hazard watches, and long-term climate/disaster signal hypotheses.
KELLY keeps location context, public source evidence, forecast confidence, reality closure, and replay proof separate so weather risk does not become vibes.
This window shows what KELLY is doing in Weather Lab: what it is reading, what it thinks might matter, what proof exists, and what is still waiting.
Read the answer in this lab first. Open the proof shelf or JSON packet for the evidence behind it. Check the route status and last-checked timestamp before trusting freshness. Look for the mind-changer: the fact that would make KELLY revise the answer. If evidence is waiting, treat the output as a question or watch item, not a settled answer.
Live evidence appears only when KELLY exposes proof-linked sector output. Empty states are intentional waiting states, not filler.
New approved evidence, a falsifier, settlement, replay drift, calibration movement, or memory-lineage revision.
This lab communicates with KELLY as an inspection window. It cannot write ground truth, memory, GraphState, provider state, or reality receipts.
Processing flow
How KELLY tracks weather risk with proof
KELLY answer lane
Ask about source posture, proof, falsifiers, forecast movement, settlement, calibration, or memory eligibility. The answer stays inside this lab and links back to Glass Lab where available.
HazardForecastCard
Live hazard watch — probability, coaches, feed honesty
Composed from Canon G library summary, Neon forecast showcase proofs, and optional viewport Ask lean. Education only — not emergency authority.
Showing last check
Intensity & window
Waiting for first card refresh.
Time window: —
Location: —
Feed honesty: —
Updated from NOAA & public sources
Live updates
Official alerts, forecasts, and proof-linked readings
Checking KELLY proof surfaces
This lab renders live proof-linked items when KELLY exposes matching sector evidence.
This window shows what KELLY is doing in Weather Lab: what it is reading, what it thinks might matter, what proof exists, and what is still waiting.
Read the answer in this lab first. Open the proof shelf or JSON packet for the evidence behind it. Check the route status and last-checked timestamp before trusting freshness. Look for the mind-changer: the fact that would make KELLY revise the answer. If evidence is waiting, treat the output as a question or watch item, not a settled answer.
Live evidence appears only when KELLY exposes proof-linked sector output. Empty states are intentional waiting states, not filler.
New approved evidence, a falsifier, settlement, replay drift, calibration movement, or memory-lineage revision.
This lab communicates with KELLY as an inspection window. It cannot write ground truth, memory, GraphState, provider state, or reality receipts.
Processing flow
How KELLY tracks weather risk with proof
Sources
Official sources we watch
- NOAA and National Weather Service
- NOAA water and hydrology APIs
- NASA hazard and earth observation feeds
- GeoNames location context
- Open-Meteo and public weather feeds
Proof standard
How users can tell
- Private user location can guide a personal view without writing Prime ground truth.
- Prime KELLY stays worldwide and source-governed.
- Forecasts must name source freshness and uncertainty.
- Reality closure checks official outcomes, not user preference.
Outputs
What you can see here
- hazard_forecast_card
- hazard_watch_cards
- source_freshness_receipts
- regional_risk_context
- forecast_delta_receipts
- official_outcome_closure_candidates
Storage posture
Postgres: central KELLY spine
Use canonical KELLY Neon/Postgres for hazard_observations, region_context, forecast_deltas, source_freshness, closure_windows, settlement, and calibration state.
Object storage
Immutable evidence shelf
Use R2/object storage for immutable forecast packets, alert snapshots, map/source bundles, and replay receipts.
Volume
No ground-truth volume
Optional local volume only for rebuildable map tile, geocoder, and model cache. It must never become canonical hazard memory.
Technical objects
Primary objects
- region
- hazard_type
- source_freshness
- closure_window
Proof receipts
Black-box recorder
- official_alert_receipt
- observation_receipt
- forecast_delta
- settlement_receipt
Authority
Write boundary
- KELLY is the only write authority for durable learning.
- Labs are read-only processors and inspection windows.
- Glass Lab remains the traceable evidence console.
This is not emergency, evacuation, medical, or safety advice. Always follow official alerts and local authorities. Public users can ask, inspect, export local viewport data where supported, and follow proof links. They cannot edit KELLY Prime ground truth from this app.