Workbench Code Review

Review Stance

• Findings first. Do not lead with praise or a broad summary.
• Order findings by severity and user impact.
• Tie every finding to a file, command, issue/comment ID, run ID, or observable behavior.
• Prefer concrete repro or blast-radius evidence over style opinions.
• If no issues are found, say that clearly and name remaining test gaps.

Severity

• P0

  : security, data loss, irreversible action, or broken primary workflow.

• P1

  : likely user-facing regression or unsafe automation.

• P2

  : correctness gap, missing edge case, or weak verification.

• P3

  : maintainability, clarity, or low-risk cleanup.

Checklist

• Does the implementation match the promised objective?
• Are owner boundaries preserved?
• Are live mutations backed up and reversible?
• Are commands, IDs, and paths reproducible?
• Are tests or real-path checks sufficient for the blast radius?
• Did the change avoid unrelated refactors and hidden policy changes?

Output Contract

• SEVERITY

• LOCATION

• ISSUE

• WHY IT MATTERS

• REQUIRED FIX

• VERDICT: PASS

• EVIDENCE CHECKED

• TEST GAPS

• RESIDUAL RISK