Interrogate the evidence

A bounded agent routes questions to local analytical tools. It does not invent measurements or depend on a runtime LLM.

Phoenix, Arizona

15 Jul 2025 · 16:00 MST

Bounded agent · no LLM

Priority zone

Tool-routing agent

Question

Which zone should we prioritize?

Agent reasoning

Intent: priority

Default priority intent; routed to hotspot ranking and environment context.

Evidence-backed answer

Shade Study Sample 07 is Priority 1 with a 74/100 multi-evidence priority score. 4.9°C of local deviation and 9 hours of persistence are material contributors.

Evidence used

42.3°C observed temperature

9 hours above the demo persistence threshold

+4.9°C from its local baseline

3,332 people/km² exposure context

Routed tools:getHotspots()getEnvironment()

The KAIRO Analyst routes questions to local deterministic tools and may recommend attention or assessment. It must not issue public-safety alerts, approve construction, or fabricate values. No tool writes to any system.

PHOENIX SCENARIO

Track 06 · Agentic AI

Interrogate the evidence

A bounded agent routes questions to local analytical tools. It does not invent measurements or depend on a runtime LLM.

Phoenix, Arizona

15 Jul 2025 · 16:00 MST

Bounded agent · no LLM

Priority zone

Tool-routing agent

Question

Which zone should we prioritize?

Agent reasoning

Intent: priority

Default priority intent; routed to hotspot ranking and environment context.

Evidence-backed answer

Shade Study Sample 07 is Priority 1 with a 74/100 multi-evidence priority score. 4.9°C of local deviation and 9 hours of persistence are material contributors.

Evidence used

42.3°C observed temperature

9 hours above the demo persistence threshold

+4.9°C from its local baseline

3,332 people/km² exposure context

Routed tools:getHotspots()getEnvironment()

Interrogate the evidence

Suggested questions

Priority zone

Interrogate the evidence

Suggested questions

Priority zone