## Security & Audits

35% weight

- Audit firm reputation and audit recency
- Verification model (optimistic, validity, multisig, MPC)
- Historical exploits, bridge incidents, and post-incident posture
- TVL relative to verifier set / multisig threshold

## Transaction Speed

20% weight

- Median time-to-fill in our hands-on test
- Worst-case time observed across 5 attempts
- Mode of operation (intent-based, lock-mint, atomic)

## Fees

20% weight

- Documented LP fee + protocol fee + gas
- True total cost on a real round-trip in our test
- Slippage observed at typical retail size ($1k, $10k)

## Chains & Tokens Coverage

15% weight

- Number of supported source / destination chains
- Token coverage on the most-used routes
- Native vs wrapped asset support

## Reliability

10% weight

- Uptime over the past 90 days (where data available)
- Frequency of route disruptions or pause events
- Front-end resilience (alternative UIs, RPC fallbacks)

Want to see how we score exchanges? [→ Exchange methodology](/content/methodology/how-we-score-crypto-exchanges/index.html)