code-review-phase-2-verify-recall-biased

Verify code change candidates using a recall-biased model.

7|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/mkusaka/ccskills --skill code-review-phase-2-verify-recall-biased
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review-phase-2-verify-recall-biased
Source: https://github.com/mkusaka/ccskills/tree/main/skills/code-review-phase-2-verify-recall-biased
Command: npx skills add https://github.com/mkusaka/ccskills --skill code-review-phase-2-verify-recall-biased

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill aids in the code review verification process by focusing on retaining plausible findings, streamlining the recall-biased verification phase.

Core Features & Use Cases

  • Recall-tier Verification: Ensures a thorough verification of candidate code changes, maintaining plausibility in the findings.
  • Bias Towards Plausibility: Keeps 'CONFIRMED' and 'PLAUSIBLE' findings, dropping 'REFUTED' findings.
  • Use Case: Ideal for automated code reviews, where maintaining accuracy and bias toward plausible findings is crucial.

Quick Start

Run the 'code-review-phase-2-verify-recall-biased' Skill to verify a batch of code change candidates.

Frequently Asked Questions about code-review-phase-2-verify-recall-biased

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify code review findings with a recall bias?

Code review verification with a recall bias evaluates candidate code changes and retains CONFIRMED and PLAUSIBLE findings while dropping REFUTED ones. This approach maximizes issue detection coverage by keeping all plausible findings during automated reviews.

What is recall-biased verification in automated code reviews?

Recall-biased verification in automated code reviews is a filtering mechanism that keeps plausible findings to avoid missing potential issues. It classifies candidates as CONFIRMED, PLAUSIBLE, or REFUTED, dropping only the refuted results to maximize recall.

How do I filter code change candidates to keep plausible findings?

To filter code change candidates and keep plausible findings, run a recall-biased verification phase that processes the batch and removes only refuted findings. This retains both confirmed and plausible results for downstream review.

When should I use a recall-biased model for code review verification?

Use a recall-biased model for code review verification when thorough issue detection is crucial in automated reviews. It is ideal when you prefer to retain all plausible findings rather than risk missing potential code defects.

What findings are kept or dropped during recall-biased code review verification?

During recall-biased code review verification, CONFIRMED and PLAUSIBLE findings are kept, while REFUTED findings are dropped. This ensures only clearly invalid candidate code changes are removed from the review pipeline.