requesting-code-review

Automate code reviews before merging by dispatching a code-reviewer subagent to assess new work using a standardized template with base and head SHAs. This provides targeted context for review and aggregates feedback by severity to guide actionable fixes.

Subagent-driven code review works by dispatching a formal code-reviewer subagent after each task or major feature. It uses precise context via placeholders and a standardized prompt to evaluate production readiness and aggregate feedback by severity.

You need to request a formal code review in your git workflow after each subagent task, after major features, and before merging to main. This prevents regressions and ensures changes align with quality standards before they cascade into bigger problems.

Yes, you can use a code review subagent for continuous engineering tasks by dispatching it after each completed task in subagent-driven development. It provides targeted context for review and aggregates structured feedback to ensure production readiness.

The best way to provide context for a code review subagent is using a standardized template that includes base and head SHAs along with precise placeholders. This ensures the subagent has targeted context to assess production readiness and return structured feedback.

You should request code reviews early in the development cycle to catch issues before they cascade into bigger problems. Early reviews ensure changes align with plan and quality standards, preventing regressions before merging to main.