Smart Confidence-Based Code Review

About

The pop-code-review skill automates the software review process by deploying parallel specialized agents to evaluate logic, architectural simplicity, and project-specific conventions. By utilizing a 80+ confidence threshold, it ensures that developers focus only on actionable, high-impact improvements rather than pedantic nitpicks or false positives. This skill is designed for major feature completion and pre-merge checks, offering a structured workflow that includes automated fixing of critical bugs, detailed quality scoring, and comprehensive re-verification of changes.

Key Features

• 0 GitHub stars
• Interactive workflow allowing for manual review, auto-fixes, and re-verification cycles.
• Intelligence-driven filtering that only reports issues with 80+ confidence scores.
• Parallel multi-agent analysis focusing on Simplicity, Correctness, and Conventions.
• Detailed assessment reports including quality scores and merge-readiness flags.
• Automated 'fix critical' mode for issues identified with 90+ certainty.

Use Cases

• Ensuring new code adheres to project-specific architectural patterns and naming conventions.
• Finalizing major feature implementations before merging into the main branch.
• Validating complex multi-step development tasks for logic errors and edge cases.