ywc-review-learnings

To stop recurring false positives in code review, you record them using a persistent Markdown file that stores project-specific review preferences. This durable memory suppresses recurring noise and enforces standards across future sessions.

Knowledge management for code review preferences works by accumulating durable rules in a version-controlled Markdown file. It records not just what to flag but why, allowing the AI to generalize rules and sharpen review quality over time.

Yes, you need the GitHub CLI installed to automate code review learning accumulation. The skill requires the GitHub CLI to harvest PR comments and integrate context-aware guidance into your review workflows.

The best way to manage DO and DO-NOT rules for code review best practices is to use polarity management within a persistent learnings file. This approach enforces standards while suppressing acceptable patterns flagged as false positives.

Yes, you can curate and update project-specific code review rules across sessions. The skill operates across read, update, list, and curate modes to maintain and sharpen the accuracy of your durable review preferences.