cyberneticlibrary

Execute exhaustive code audit with verification

code-review-deepworkflowsetup L30
Cloud-Officer/claude-code-plugin-dev
What it does

Deep code review with multi-agent correctness, efficiency, and architecture lenses

Best for

Review code when multi-lens analysis (correctness + design) and configurable effort are required.

Inputs
  • · diff/branch
  • · effort level (low/medium/high/ultra)
  • · focus area (optional)
Outputs
  • · structured findings across correctness/reuse/efficiency + inline PR comments (optional)
Requires
  • · git (diff parsing)
  • · static analysis (optional)
Preconditions
  • · diff is coherent (not massive churn)
  • · codebase conventions are clear
  • · review agents available
Failure modes
  • · findings are shallow (style, not substance)
  • · false positive bugs (code is correct)
  • · review is incomplete (effort mismatch)
Trust signals
  • · effort level matched to finding count
  • · file:line specificity
  • · proposed fixes are sound