Scoring Engine

Complete reference for all scoring algorithms, formulas, and thresholds.

24,502

Star Systems

scored per cycle

Trust Dimensions

per player

CHI Sub-Indices

weighted composite

Alert Types

anomaly patterns

Consistency Guarantee

The same scoring formulas run in three places: the oracle backend (scoring.ts), the frontend live data module (liveData.ts), and the Sui smart contract (frontier_pulse.move). This ensures consistent scores regardless of which path produces them.

System Health Scoring

Each of the 24,502 star systems receives a health snapshot. When real enrichment data (smart assemblies, killmails) is available from the World API, it drives the scores. Otherwise, a deterministic hash function provides consistent fallback scores.

Real Data Path

Inputs from World API enrichment:

Input	Source	Description
playerCount	Smart Assemblies + Killmails	Unique player addresses active in system
infraCount	Smart Assemblies API	Number of deployed Smart Assemblies (gates, SSUs, turrets)
kills	Killmails API	Recent PvP kill count in this system

scoring.ts — System Health

// Step 1: Component scores (0-100 each)
playerScore  = min(playerCount * 5, 100)
infraScore   = min(infraCount * 3, 100)
combatScore  = min(kills * 8, 100)

// Step 2: Activity Level — weighted blend
activityLevel = clamp(
  playerScore * 0.4 + infraScore * 0.35 + combatScore * 0.25
)

// Step 3: Trust Level — inverse of combat ratio + infrastructure bonus
combatRatio = playerCount > 0 ? kills / playerCount : 0
baseTrust   = clamp(100 - combatRatio * 50)
infraBoost  = min(infraCount * 2, 20)
trustLevel  = clamp(baseTrust + infraBoost)

// Step 4: Transaction Frequency — activity density proxy
txFrequency = clamp((playerCount * 3 + infraCount * 2 + kills) * 2)

// Step 5: Local CHI — 40% activity + 60% trust (trust-weighted)
localChi = floor((activityLevel * 40 + trustLevel * 60) / 100)

Activity Level

players×0.4 + infra×0.35 + combat×0.25

Trust Level

100 - combatRatio×50 + infraBoost

Local CHI

activity×40% + trust×60%

Deterministic Fallback

When no enrichment data exists, a seeded hash function generates consistent pseudo-random scores from the system ID:

vitals.ts — Hash Function

function hash(n: number): number {
  let h = n ^ 0x5f3759df;
  h = Math.imul(h ^ (h >>> 16), 0x45d9f3b);
  h = Math.imul(h ^ (h >>> 13), 0x45d9f3b);
  h = (h ^ (h >>> 16)) >>> 0;
  return (h & 0xffff) / 0xffff;  // → 0.0 to 1.0
}

// Same system ID always produces the same vitals:
activityLevel = floor(10 + hash(systemId) * 90)           // 10-100
trustLevel    = floor(15 + hash(systemId + 7919) * 85)    // 15-100
playerCount   = floor(hash(systemId + 15373) * 60)        // 0-60
infraCount    = floor(hash(systemId + 23197) * 20)        // 0-20
txFrequency   = floor(10 + hash(systemId + 31531) * 90)   // 10-100
combatIncidents = floor(hash(systemId + 40343) * 12)      // 0-12

The hash offsets (7919, 15373, etc.) are prime numbers chosen to minimize correlation between dimensions for the same system.

Player Reputation (Trust Compass)

Each player's on-chain activity is analyzed across 5 behavioral dimensions (0-100 each):

scoring.ts — Player Reputation

// Inputs from World API aggregation:
//   assemblyCount  — Smart Assemblies owned by this player
//   killCount      — Times this player was the attacker
//   deathCount     — Times this player was the victim
//   systemsVisited — Unique systems with activity

totalCombat     = killCount + deathCount
aggressionRatio = totalCombat > 0 ? killCount / totalCombat : 0

// 5 Dimensions:
reliability  = clamp(40 + assemblyCount * 5 + systemsVisited * 2)
commerce     = clamp(30 + assemblyCount * 8)
diplomacy    = clamp(50 - aggressionRatio * 40 + systemsVisited * 3)
stewardship  = clamp(20 + assemblyCount * 10)
volatility   = clamp(totalCombat * 5 + |killCount - deathCount| * 3)

Trust Compass Dimension Weights

25%

20%

10%

Reliability (25%)

Commerce (25%)

Diplomacy (20%)

Stewardship (20%)

Inv. Volatility (10%)

Composite Trust Score

Weighted combination matching the on-chain formula:

On-chain formula

composite = (
  reliability * 25 +
  commerce    * 25 +
  diplomacy   * 20 +
  stewardship * 20 +
  (100 - volatility) * 10   // inverted: low volatility = good
) / 100

Archetype Classification

First matching rule wins:

Archetype	Rule	Description
Civilization Builder	stewardship ≥ 80 AND reliability ≥ 70	Builds infrastructure for others
Trusted Trader	commerce ≥ 80 AND reliability ≥ 70	Reliable economic participant
Diplomat	diplomacy ≥ 75 AND volatility < 30	Brings people together peacefully
Warlord	volatility ≥ 70 AND commerce < 40	Combat-focused, low trade
Wildcard	50 ≤ volatility < 70	Unpredictable behavior
Newcomer	default	Insufficient data for classification

Civilization Health Index (CHI)

The global CHI aggregates all system health scores into a single composite (0-100) using 6 weighted sub-indices:

CHI Sub-Index Weights

20%

15%

20%

15%

Economic Vitality (20%)

Security Index (15%)

Growth Rate (15%)

Connectivity (15%)

Trust Index (20%)

Social Cohesion (15%)

scoring.ts — Global CHI

// Inputs: all system health scores from current cycle

economicVitality = avg(txFrequency) * 0.6 + avg(infraCount * 5) * 0.4
securityIndex    = clamp(100 - avg(combatIncidents) * 8)
growthRate       = (systemsWithActivity>50 / totalSystems) * 100
connectivity     = avg(activityLevel) * 1.1
trustIndex       = avg(trustLevel)
socialCohesion   = trustIndex * 0.4
                 + securityIndex * 0.3
                 + min(avg(playerCount) * 3, 100) * 0.3

// Weighted overall (matches on-chain):
overall = (
  economicVitality * 20 +
  securityIndex    * 15 +
  growthRate       * 15 +
  connectivity     * 15 +
  trustIndex       * 20 +
  socialCohesion   * 15
) / 100

Diagnosis Thresholds

Score	Diagnosis	Meaning
≥ 80	Flourishing	Peak civilization — high activity, high trust, growing
≥ 65	Thriving	Strong and healthy, minor issues
≥ 50	Stable	Functional but not exceptional
≥ 35	Stressed	Cracks showing — trust or security concerns
≥ 20	Declining	Significant deterioration across metrics
< 20	Collapsing	Civilization is failing

Anomaly Detection

Each oracle cycle scans system scores against pattern rules. Alerts are emitted as on-chain events (not stored, for gas efficiency).

Alert Type	Trigger Condition	Severity	On-Chain Code
Blackout	infra > 5 AND activity < 10	Critical (0)	AnomalyAlertEmitted
Trust Collapse	trust < 20 AND players > 5	High (1)	AnomalyAlertEmitted
Combat Hotspot	combatIncidents > 8	Medium (2)	AnomalyAlertEmitted
Trade Spike	tx > 85 AND players > 20	Warning (3)	AnomalyAlertEmitted
Infrastructure Hub	infraCount > 5	Info (4)	AnomalyAlertEmitted

The frontend's liveData.ts also generates additional summary alerts (Population Center, Warzone, Civilization Core, Census Update) from aggregate statistics.

Color Mapping

Trust scores map to a consistent color palette across all visualizations:

Trust Level Spectrum

0Hostile40Stressed70Healthy100

lib/colors.ts

// Trust colors (galaxy nodes, compass, badges)
trust >= 70 → green  (#00ff88)  // Healthy
trust >= 40 → orange (#ff9800)  // Stressed
trust <  40 → red    (#ff3d3d)  // Hostile

// CHI gauge colors
score >= 70 → green  (#00ff88)
score >= 50 → cyan   (#00e5ff)
score >= 30 → orange (#ff9800)
score <  30 → red    (#ff3d3d)

// Alert severity colors
critical → #ff3d3d  |  high → #ff6b35  |  medium → #ff9800
warning  → #ffca28  |  info → #00e5ff