HomeDocsIntroduction

Solana Intelligence Engine

ZEBRA Documentation

Real-time blockchain intelligence for detecting anomalies, tracking whale movements, and identifying rug pulls on Solana.

1.2B+

TXs Analyzed

99.7%

Detection Rate

<50ms

Alert Latency

1Quick Start

Terminal Commands

scan <address>Analyze wallet for risk factors

pattern <address>Visualize transaction patterns

whalesTrack large wallet movements

rugsView rug pull detections

chaosMarket volatility metrics

feedLive transaction feed

2Architecture

Ingestion

RPC, Geyser

Processing

Pattern ML

Risk Engine

Multi-factor

Alerts

WS, Hooks

ZEBRA processes ~4,000 TPS from Solana mainnet through 47 detection heuristics, maintaining a 24h rolling cache of 350M+ transactions for historical analysis.

3Pattern Recognition

Wash Trading99.2%

Circular patterns via graph analysis

Sybil Detection97.8%

Coordinated wallet creation patterns

Liquidity Extraction98.5%

Abnormal DEX withdrawal patterns

Mint Authority99.9%

Unauthorized token creation events

4Risk Scoring

Weighted Ensemble (12 factors)

Behavioral Clustering20%ML similarity to malicious archetypes

Funding Source18%Mixer detection, CEX patterns

Wallet Age15%Time since first TX, activity consistency

Liquidity Risk15%Holdings in suspicious tokens

TX Diversity12%Counterparty entropy

Network Centrality10%Graph-based role analysis

Temporal Patterns10%Bot-like timing signatures

5Whale Tracking

Whale

>$10M

Dolphin

$1M-$10M

Fish

$100K-$1M

Real-time index of 2,847 whale wallets. Alerts trigger at 0.5% portfolio movement with 2.3s average detection latency.

6Rug Detection

Pre-Rug Indicators (23 monitored)

LP concentration shifts

Abnormal mint activity

Dev wallet funding changes

Contract upgrade retention

Coordinated sell pressure

Fake volume patterns

4.2 min median pre-rug alert

78% of users acting on alerts exited with <15% loss vs 95%+ eventual decline.

7WebSocket API

Connectionwss://api.zebra.io/v1/stream

alerts.rugRug pull alerts

alerts.whaleWhale movements

metrics.chaosChaos index (1s)

tx.suspiciousFlagged TXs

8REST Endpoints

GET/v1/scan/{address}Wallet risk analysis

GET/v1/whalesTracked whale list

GET/v1/rugsRecent detections

GET/v1/chaosMarket metrics

POST/v1/alerts/subscribeConfigure webhooks

9Rate Limits

Free100/min1 conn50/day

Pro1,000/min5 connUnlimited

Enterprise10,000/min50 connUnlimited

10Authentication

Include your API key in the Authorization header:

Authorization: Bearer <your_api_key>

11Custom Scoring

ZEBRA allows you to create custom scoring models by combining and weighting different risk factors according to your trading strategy.

Scoring Model Configuration

{
  "model_name": "aggressive_rug_detector",
  "weights": {
    "liquidity_score": 0.35,
    "holder_distribution": 0.25,
    "contract_analysis": 0.20,
    "social_sentiment": 0.10,
    "volume_anomaly": 0.10
  },
  "thresholds": {
    "high_risk": 0.75,
    "medium_risk": 0.50,
    "low_risk": 0.25
  },
  "decay_rate": 0.95
}

Available Factors

liquidity_score
holder_distribution
contract_analysis
social_sentiment
volume_anomaly
whale_correlation

Decay Functions

linear - Constant decay
exponential - Fast initial decay
logarithmic - Slow initial decay
step - Discrete thresholds

Create Custom Model

POST /api/v1/models/custom

12Alert Webhooks

Configure real-time webhook notifications for critical events. Webhooks are delivered with HMAC-SHA256 signatures for verification.

Webhook Configuration

{
  "endpoint": "https://your-server.com/webhook",
  "secret": "whsec_...",
  "events": [
    "rug.detected",
    "whale.movement",
    "liquidity.removed",
    "score.critical"
  ],
  "filters": {
    "min_value_sol": 100,
    "risk_threshold": 0.7
  },
  "retry_policy": {
    "max_attempts": 3,
    "backoff": "exponential"
  }
}

Event Types

rug.detectedPotential rug pull identified

rug.confirmedRug pull confirmed on-chain

whale.buyLarge accumulation detected

whale.sellLarge distribution detected

liquidity.addedLP tokens minted

liquidity.removedLP tokens burned

score.criticalRisk score exceeds threshold

score.changeSignificant score movement

Webhook Payload

{
  "id": "evt_1a2b3c4d5e6f",
  "type": "rug.detected",
  "timestamp": 1703894400000,
  "data": {
    "token": "So11111111111111111111111111111111111111112",
    "risk_score": 0.92,
    "indicators": ["lp_unlock", "dev_wallet_active"],
    "estimated_impact": "450000 SOL"
  },
  "signature": "sha256=..."
}

13Data Export

Export historical analysis data, patterns, and scoring records for offline analysis, compliance reporting, or integration with external systems.

JSON

.json

Structured data

CSV

.csv

Spreadsheet compatible

Parquet

.parquet

Columnar analytics

Export Request

{
  "type": "historical_scores",
  "format": "parquet",
  "date_range": {
    "start": "2024-01-01T00:00:00Z",
    "end": "2024-01-31T23:59:59Z"
  },
  "filters": {
    "tokens": ["*"],
    "min_risk_score": 0.5,
    "include_metadata": true
  },
  "compression": "gzip",
  "delivery": {
    "method": "s3",
    "bucket": "your-bucket",
    "prefix": "zebra-exports/"
  }
}

Available Datasets

historical_scoresComplete scoring history for all tracked tokens

~2GB/month

whale_movementsAll detected whale transactions and patterns

~500MB/month

rug_incidentsConfirmed and suspected rug pull events

~50MB/month

pattern_matchesPattern recognition results and confidence scores

~1GB/month

market_chaosChaos index values and contributing factors

~100MB/month

Delivery Methods

Direct download (up to 1GB)
Amazon S3 bucket
Google Cloud Storage
SFTP endpoint

Scheduling

One-time export
Daily incremental
Weekly full snapshot
Custom cron schedule