Manipulation Detection
Powered by Allium, the Manipulation Detection dashboard identifies artificial trading patterns that may indicate market manipulation.
Patterns Detected
| Pattern | What It Means | Confidence |
|---|---|---|
| Wallet Clustering | Multiple wallets controlled by the same entity trading in coordination | High |
| Interval Regularity | Trades happening at suspiciously regular intervals (bot-like timing) | Medium |
| Size Distribution | Identical or near-identical trade amounts (not natural) | Medium |
| Coordinated Timing | Multiple wallets executing trades at the same time | High |
| New Wallet Spam | Fresh wallets with no history suddenly active on a token | Medium |
| Circular Trading | SOL/tokens flowing in loops between related wallets | High |
| Wash Trading | Same entity buying and selling to inflate volume | High |
How to Use
- Navigate to the Detection tab
- Enter a token CA to analyze
- Click Scan
- Review detected patterns with confidence scores
Alert Output
Each detection returns:
- Priority: Low / Medium / High / Critical
- Patterns Found: List of detected manipulation types
- Confidence Score: How certain the detection is (0-100%)
- Evidence: Specific wallet addresses and transactions involved
- Recommendation: Whether to avoid the token
Data Source
Detection analysis uses the Allium API which indexes Solana transaction data and applies machine learning models to identify manipulation patterns. The analysis considers:
- Transaction timing and intervals
- Wallet creation dates and history
- Fund flow between wallets
- Trade size distributions
- Network graph analysis of related wallets
WARNING
Detection is probabilistic, not definitive. A "clean" scan doesn't guarantee safety — it means no known manipulation patterns were detected. Always combine with Shield Scanner results and your own research.