What problem does it solve? Code reviews of fast or AI-assisted Go changes often miss structural violations, duplicated concerns, concurrency leaks, and untested behavior. This Skill enforces gokit's standing engineering baseline as a repeatable, independent review so defects in shared foundation code are caught before merge or release. ## Core Features & Use Cases - Eight focused review passes: structure & placement, canonical reuse, principles, security & privacy, quality, tests/TDD, docs & supply chain, and comments & godoc, each with detection commands and severity rules. - Three orchestration drivers: review a change set (branch, commit range, or HEAD~1), audit a whole module or tree, or fan the review into parallel subagent passes that plan and apply fixes. - Severity-classified findings: every issue is reported as blocker, should-fix, or nit with file:line, the violated principle, and a suggested fix. - Use Case: Before merging a branch that touches gokit's resilience package, dispatch this Skill to a fresh reviewer agent to verify layering, canonical reuse, race-clean tests, and coverage gates, then receive a consolidated findings report. ## Quick Start Ask the AI to run the review skill on the current branch diff against origin/main and report all blocker and should-fix findings.