What problem does it solve?
It prevents defects and misaligned implementations from propagating by ensuring changes are reviewed against requirements before teams merge or proceed to the next step.
Core Features & Use Cases
- Review-trigger guidance: Defines mandatory and optional moments to request review across subagent-driven development, plan execution, and ad-hoc changes.
- Deterministic review dispatch: Standardizes how to compute git base/head SHAs and how to submit a structured request to a code-reviewer subagent using a reusable template.
- Feedback-driven iteration: Instructs how to act on Critical/Important/Minor findings and how to push back when feedback is incorrect, using technical justification.
Example use case: after completing a large feature, compute the BASE_SHA and HEAD_SHA for that change, dispatch the code-reviewer subagent with a clear summary and plan link, then fix any Critical and Important issues before merging.
Quick Start
Ask the AI to request a code review for the current implementation by providing BASE_SHA and HEAD_SHA and filling the code-reviewer template with your change summary and plan requirements.