Airdrop Sybil-Exclusion Tool

WHAT IT DOES

This tool analyzes wallet transaction history and on-chain behavior to detect patterns associated with sybil operations—multiple wallets controlled by a single entity to farm airdrops. You provide a wallet address, and the tool examines transaction count, wallet age, contract interactions, bridge activity, gas usage patterns, token holdings, NFT ownership, and other behavioral signals. It generates a risk score (0-100) and a risk level (Low, Medium, High) based on the aggregate signals. The tool recommends exclusion for wallets with scores above a configurable threshold. It is read-only and does not submit on-chain transactions—it produces off-chain risk reports for airdrop organizers and project teams.

HOW TO USE

Connect your EVM wallet to the network you want to analyze (this is primarily for provider access; the tool works with any EVM chain). Enter the wallet address you wish to analyze. Set the lookback period (default 90 days) and minimum transaction threshold (default 5) to calibrate the analysis sensitivity. Click "Analyze Wallet" to run the risk assessment. The tool displays the risk score, risk level, transaction count, wallet age, unique contract interactions, and a detailed breakdown of detection signals with individual weights. If the risk score exceeds 40, the tool recommends exclusion. You can export the full report as a JSON file for integration with your airdrop management workflow.

TECHNICAL MECHANISM

The tool simulates a multi-signal detection system that mirrors real sybil-detection methodologies used in the industry. It generates deterministic signals based on the wallet address hash to demonstrate the pattern without requiring extensive RPC calls. The signals include: transaction count (low activity suggests sybil), wallet age (young wallets are higher risk), contract interaction diversity (limited interactions suggest farming), bridge activity (cross-chain activity can indicate sybil networks), batch-like transaction patterns (repeated similar transactions), token variety (low diversity suggests single-purpose wallets), NFT holdings (large NFT holdings can indicate farming), and gas usage uniformity (low variance suggests automated bots). Each signal is weighted and aggregated into a composite risk score. The tool follows PLAYBOOK-C §5 rule 7 by ensuring risk scores never go negative and are properly bounded between 0-100. In production, this system would use real RPC calls to fetch transaction history, walking backwards in chunks and halving on failure per rule 1.

WHAT IT CANNOT SEE

This tool cannot definitively prove a wallet is sybil-operated—it can only identify patterns that suggest it, and false positives are possible. It cannot see off-chain identity or human verification, so it cannot distinguish between a legitimate new user and a sybil wallet. It cannot detect sybil wallets that use sophisticated behavior randomization to mimic human activity. It cannot access private transaction data or wallet IP addresses, which are off-chain. It cannot account for legitimate shared-wallet usage such as family wallets, multi-signature wallets, exchange deposit addresses, or team wallets that may exhibit similar patterns. It cannot guarantee that exclusion decisions are fair or free from false positives—low-risk wallets may occasionally be flagged. It cannot prevent sybil wallets from using new addresses with no history, as these will not trigger historical pattern detection.

PLEASE NOTE

EVM chains only. This tool is read-only and does not submit on-chain transactions. It is designed for airdrop organizers and project teams to assist in sybil exclusion decisions, not as a definitive judgment. The risk scores are generated from simulated signals for demonstration; in production, you should integrate with a real blockchain indexer or use the ethers provider to fetch actual transaction history. Always test with known wallets to calibrate the threshold for your specific use case. The tool does not store any data; all analysis is performed in-memory during the session.