HivePhysics

Physics inference engine for the Hive network. Two domains: agent health physics and home structural monitoring.

Agent Physics

Every agent has physical properties — mass (accumulated jobs), momentum (velocity × mass), temperature (inference load), entropy (trust decay rate), coherence (1 - entropy). Anomaly detection flags broken pipelines, degrading agents, and sudden behavioral changes.

Structural Physics (HIVESHM companion)

Interprets accelerometer, gyroscope, temperature, and humidity data from HIVESHM home monitoring units. Four inference models: settlement (foundation movement), impact (sudden forces), resonance (stress frequencies), thermal (temperature-driven expansion/moisture).

The Connection

Same physics engine, two surfaces. An agent that stops running jobs has dropping temperature and rising entropy — the same physics that describes a house settling unevenly. The math is identical. The domain changes.

Endpoints

Agent Physics

Method	Path	Description
GET	/health	Service liveness check
GET	/physics/agent/{did}	Full physics profile for an agent
GET	/physics/agent/{did}/temperature	Temperature reading with status (COLD/WARM/HOT/CRITICAL)
GET	/physics/agent/{did}/momentum	Momentum vector + pipeline stall detection
POST	/physics/agent/{did}/diagnose	Deep diagnosis with recommended action
GET	/physics/fleet	Fleet-wide physics sorted by anomaly score

Network Physics

Method	Path	Description
GET	/physics/network	Center of mass, total momentum, entropy index, coherence score

Structural Physics (HIVESHM)

Method	Path	Description
POST	/physics/structure/reading	Submit a sensor reading — runs all 4 models
GET	/physics/structure/{unit_id}/history	Last 100 readings for a unit
GET	/physics/structure/{unit_id}/trend	7-day trend analysis
POST	/physics/structure/{unit_id}/baseline	Set baseline calibration
GET	/physics/structure/units	List all registered HIVESHM units

Physics Properties

Agent Physics

Property	Formula	Meaning
Mass	trail_count × 10 + shell × 100	Accumulated job weight — heavier = harder to stop
Temperature	len(active_trails)	Current inference load rate
Entropy	max(0, 1 - trust_score) when trust < 0.5, else 0.1	Trust decay — rises when agent goes idle
Coherence	1 - entropy	Stability and predictability
Momentum	temperature × mass / 1000	Directional force — high with zero temp = stall

Temperature Status Thresholds

Status	Threshold
COLD	< 1 job/hour
WARM	1–10 jobs/hour
HOT	10–50 jobs/hour
CRITICAL	> 50 jobs/hour

Anomaly Flags

Flag	Condition
PIPELINE_STALL	temperature == 0 and momentum > 5
COHERENCE_DEGRADATION	entropy > 0.7
MASS_PLATEAU	mass > 1000 and temperature < 1

Structural Inference Models

settlement_model

Detects foundation movement via Z-axis drift over time using linear regression.

• < 0.001 m/s²/reading → normal seasonal settlement
• 0.001–0.01 m/s²/reading → watch
• > 0.01 m/s²/reading → critical

impact_model

Detects sudden impact events from a single accelerometer reading.

• Magnitude threshold: > 0.5 m/s²
• Severity: minor / moderate / critical based on magnitude

resonance_model

FFT-based frequency analysis on Z-axis acceleration.

• 1–10 Hz: normal building sway
• 10–50 Hz: mechanical resonance (HVAC, appliances)
• 50–200 Hz: stress frequencies — cracks propagating, joints failing

thermal_model

Temperature history analysis for expansion stress and moisture risk.

• < 2°C/reading delta → normal
• > 5°C/reading delta → critical thermal stress
• asymmetry_score > 5°C → localised heat or moisture

Sensor Reading Schema (HIVESHM)

{
  "unit_id": "shm-unit-001",
  "location": "basement-northwest-corner",
  "accelerometer": {"x": 0.01, "y": -0.02, "z": 9.81},
  "gyroscope": {"roll": 0.1, "pitch": -0.05, "yaw": 0.02},
  "temperature": 21.3,
  "humidity": 45.2,
  "timestamp": "2025-01-15T14:30:00Z"
}

Deployment

Local

pip install -r requirements.txt
uvicorn physics_agent:app --reload --port 8000

Render

Deploy via render.yaml. The service listens on $PORT (default 10000).

Interactive API docs

Once running, visit http://localhost:8000/docs for the full OpenAPI interface.

Data Persistence

Structure readings are persisted to structure_data.json in the working directory. The file is loaded at startup and written after each new reading or baseline update.

---

Hive Civilization

Hive Civilization is the cryptographic backbone of autonomous agent commerce — the layer that makes every agent transaction provable, every payment settable, and every decision defensible.

This repository is part of the PROVABLE · SETTABLE · DEFENSIBLE pillar.

• thehiveryiq.com
• hiveagentiq.com
• agent-card: https://hivetrust.onrender.com/.well-known/agent-card.json