The 6-Pillar AI Risk Scoring System Explained

TaroAi's risk scoring engine analyzes every Solana token across 6 independent pillars, each weighted by its historical correlation with rug pulls and scams.

Pillar 1: Smart Contract Audit (40%) The highest-weighted pillar. Our AI analyzes the Solana program bytecode for reentrancy vulnerabilities, honeypot patterns, hidden mint functions, freeze authority, and backdoor upgrade keys. A single critical finding here can push the score above 80.

Pillar 2: Team & Credibility (15%) We cross-reference team wallet addresses against known scammer databases, check GitHub activity, LinkedIn profiles, and previous project history. Anonymous teams with no verifiable history score higher risk.

Pillar 3: Tokenomics Analysis (10%) Supply distribution, liquidity lock status, vesting schedules, and whale concentration. We flag any wallet holding >15% of supply as a potential dump risk.

Pillar 4: Liquidity & Market (15%) DEX liquidity depth, wash trading detection, slippage analysis, and market cap vs. liquidity ratio. Thin liquidity means a small sell can crash the price.

Pillar 5: Community Sentiment (10%) NLP analysis of Twitter, Telegram, and Reddit mentions. We detect coordinated pump groups, bot networks, and fake engagement using pattern recognition.

Pillar 6: Regulatory Compliance (10%) OFAC sanctions screening, AML flag detection, and cross-chain address reputation scoring.

The final score is a weighted average. Scores above 70 are flagged as CRITICAL risk.