BlackHartBlackHart
Scores/Methodology/Adversarial Resilience
10% weight

Dimension 7: Adversarial Resilience

Measures protocol resilience under independent adversarial security research. Higher scores reflect stronger resistance to sophisticated attack methodologies.

What We Measure

We measure how well a protocol withstands independent, offensive security research conducted through advanced adversarial methodologies. Unlike other dimensions that analyze static properties of the codebase, Adversarial Resilience reflects the outcome of actively attempting to compromise the protocol using sophisticated attack techniques spanning economic exploits, cross-contract composition attacks, and novel vulnerability classes. The score represents the protocol's demonstrated resistance to real attack pressure — not theoretical analysis, but actual adversarial engagement with the live system.

What Raises This Score

+

Strong resistance demonstrated across all tested attack surfaces

+

Protocol architecture that inherently limits adversarial options

+

All previously identified areas of concern fully remediated

+

Deep testing coverage across multiple attack methodology categories

+

Active, well-funded bounty program incentivizing continuous external scrutiny

+

Architecture designed with adversarial assumptions from the ground up

+

Clean record across multiple independent adversarial assessments

What Lowers This Score

-

Areas of concern identified during adversarial testing

-

Novel attack surfaces that have not yet been fully stress-tested

-

Limited adversarial testing history or narrow coverage

-

Architecture that creates complex interaction paths difficult to reason about

-

Evolving codebase requiring continuous re-assessment

-

Attack surfaces that are theoretically bounded but practically unexplored

-

Low bounty coverage relative to TVL-at-risk

Why This Weight

At 10%, Adversarial Resilience reflects that real-world attack resistance is the ultimate test of security — but it requires other dimensions to be strong first. A protocol cannot be adversarially resilient if its access control is broken or its economics are unsound. This dimension captures the residual risk that remains after all structural defenses are in place: can an intelligent adversary still find a way through?