sybil-detection
Sybil Detection — Coordinated Wallet & Fake Activity Analysis
Sybil attacks in Solana token markets involve a single entity operating many wallets to create the illusion of organic activity. This skill covers detecting coordinated wallet clusters, wash trading, bundled transactions, and fake holder inflation — critical for evaluating whether a token's metrics reflect real demand or manufactured signals.
Why Sybil Detection Matters
Token markets on Solana are rife with manufactured signals:
- Inflated holder counts: 500 "holders" that are really 10 entities with 50 wallets each
- Fake volume: Wash trading between self-controlled wallets to simulate demand
- Artificial social proof: Many wallets holding small amounts to appear broadly distributed
- Rug preparation: Creator distributes supply across many wallets, then sells coordinated
- Bundled launches: PumpFun tokens where creator buys via Jito bundle in first slot
A token showing 1,000 holders with 80% funded from 3 wallets is fundamentally different from one with 1,000 independently-funded holders. Sybil detection separates real demand from theater.
Detection Categories
1. Funding Source Analysis
More from agiprolabs/claude-trading-skills
pandas-ta
Technical analysis with 130+ indicators using pandas-ta for crypto market data
105risk-management
Portfolio-level risk controls, drawdown management, exposure limits, and circuit breakers for crypto trading
76feature-engineering
Feature construction from market data for ML trading models including price, volume, on-chain, and microstructure features
76trading-visualization
Professional trading charts including candlesticks, equity curves, drawdowns, correlation heatmaps, and return distributions
76signal-classification
ML trading signal classifiers using XGBoost and LightGBM with walk-forward validation, SHAP feature importance, and threshold optimization
73market-microstructure
DEX orderflow analysis, trade classification, buyer/seller pressure, and microstructure signals for Solana tokens
73