What problem does it solve? Raw severity scores alone do not reflect what actually matters on a given machine: the same finding ranks differently on a public server than on a workstation, and a regression deserves more attention than a long-standing known issue. This Skill orders the findings journal so the report leads with what matters on this specific host. ## Core Features & Use Cases - Context-aware scoring: Scores each open or regressed finding using base severity, profile weight via profile_severity, status, category, and exposure (public server vs workstation). - Severity normalization: Updates a finding's severity in the SQLite journal when the machine profile assigns it a different weight than the default, keeping reports and thresholds consistent. - Regression boost: Elevates regressed findings, since a fix that came back is urgent. - Use Case: After correlate-findings computes the delta on a public web server, run this Skill so report-status and send-report lead with the regressed SSH hardening finding instead of a stale low-risk notice. ## Quick Start Ask the AI to run the prioritize-redflags skill after correlate-findings to rank the journal findings by what matters most on this machine before generating the report.
