Severity Classification Skill

Classify audit findings by severity using data-driven benchmarks and historical distributions calibrated from major firms. This Skill applies these distributions across HIGH, MEDIUM, LOW, and GAS levels to return a labeled severity with traceable justification.

Data-driven vulnerability classification supports HIGH, MEDIUM, LOW, and GAS severity levels. Each level includes 30 representative examples of typical vulnerabilities and narratives to ensure consistent risk assessment across diverse vulnerability types.

Generate auditable severity justifications by applying historical severity distributions to your audit findings. The classification process outputs a labeled severity alongside a scoring weight reference to ensure traceable and consistent risk assessment for internal reports.

Yes, you can standardize audit severity classifications using historical calibration data from Code4rena, Sherlock, Cyfrin, and other firms. These benchmarks provide the data-driven distributions necessary to consistently assign severity levels to your audit findings.

The best way to apply consistent risk assessment is using a data-driven severity classification approach. By leveraging historical distributions and representative examples per vulnerability level, teams ensure consistent and traceable severity labeling across diverse smart contract findings.

You need historical calibration data to classify audit findings by severity because it provides the benchmarks and historical distributions required. This Skill includes pre-installed references with 30 examples per level to ensure accurate and consistent risk assessment.