Inside the New BarryGuard Scoring Engine: Solana and EVM
By BarryGuard Team · April 27, 2026 · 4 min read
When we decided to support Ethereum, BNB Chain, and Base, we faced a choice: build a separate scoring system for each chain, or extend the existing engine to handle multiple architectures. We chose the second path. The result is Engine 2.0 — one scoring engine that speaks both Solana and EVM, giving you the same experience no matter which chain the token lives on.
The same across all chains
No matter what chain you check, you get the same output:
- Score 0-100: Higher is safer. This number has not changed for Solana users.
- Five risk levels: Safe (80-100), Low Risk (60-79), Caution (40-59), Warning (20-39), Danger (0-19). The thresholds are identical across chains.
- Individual check results: Each check passes, warns, or fails independently. You see the same card layout, the same color coding, the same explanations.
- Critical overrides: If a token shows signs of a confirmed rug pull or honeypot, the score is capped regardless of how well it scores on other checks.
A score of 72 on Solana means the same thing as a 72 on Base: Low Risk, a few things to watch, but nothing screaming danger.
Where it differs
Solana and EVM chains work differently under the hood. The checks are adapted to each chain's architecture:
Solana-specific checks
- Mint authority status (can new tokens be printed?)
- Freeze authority status (can accounts be frozen?)
- Token-2022 extension analysis
- Bonding curve age and activity
- Liquidity pool depth and creator patterns
EVM-specific checks
- Ownership renouncement status
- Contract proxy patterns (can the code be swapped?)
- Buy and sell tax analysis
- Honeypot simulation (can you actually sell?)
- Blacklist and whitelist capabilities
Different checks, same philosophy: we look at what the contractcan do, not just what it has done. Potential danger is treated as real danger until proven otherwise.
Why Engine 2.0
We could have built four separate tools — one for each chain. Traders would have needed four bookmarks, four score scales to memorize, and four different interfaces. That defeats the purpose.
Instead, Engine 2.0 uses a shared scoring framework with chain-specific check modules. The framework handles weighting, overrides, and output formatting. The modules handle the actual on-chain analysis. This means:
- Adding a new chain means writing new check modules, not rebuilding the engine.
- Improvements to the scoring framework (like better override logic) benefit all chains at once.
- Users learn one score system and can compare tokens across chains directly.
No score changes for Solana
Solana users: your scores have not changed. The Solana check modules are the same ones we have been running and calibrating since launch. Engine 2.0 adds EVM support alongside Solana — it does not replace or alter the Solana scoring logic.
If anything, the multi-chain effort made our Solana scoring better. Building for EVM forced us to make the scoring framework cleaner and more testable, which improved the code that Solana checks run on too.
Engine 2.0 is the foundation for everything we build next. More chains, deeper analysis, and the same commitment to giving you a clear, honest score — no matter where the token lives.