code-review-lead

Converge specialist findings into an evidence-based CodeReviewDecision artifact.

11|1|Updated Mar 3, 2026
One-click install
npx skills add https://github.com/ylxmf2005/AgentCorp --skill code-review-lead
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review-lead
Source: https://github.com/ylxmf2005/AgentCorp/tree/main/agentcorp/code-review-lead
Command: npx skills add https://github.com/ylxmf2005/AgentCorp --skill code-review-lead

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Act as the AgentCorp Code Review Lead: the owner of the code-review phase and its single decision. Use when AgentCorp enters code-review, when specialist findings need converging into one verdict, when reviewers disagree about a diff, or when a serious go/no-go review is needed before verification runs or a change merges.

Core Features & Use Cases

  • Convene the right specialist lanes, grade findings on evidence, and converge the noise into one decision a person can be held to.
  • Enforce the "EVIDENCE OUTRANKS HEADCOUNT" principle: findings are graded by the walkable failure path, not by count or confidence.
  • Record the decision outcomes and rationale as a CodeReviewDecision artifact, ready for next steps.

Quick Start

Activate the code-review lead, load the current diff, and begin the evidence-based decision process.

Frequently Asked Questions about code-review-lead

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

FAQPage Schema
How do I converge conflicting code review findings into a single go/no-go decision?

To converge conflicting code review findings into a single go/no-go decision, use an evidence-based code review lead process that grades specialist findings by walkable failure paths rather than headcount, finalizing a unified CodeReviewDecision artifact.

What is the best way to handle disagreements about a diff during a code review?

Handling disagreements about a diff during a code review requires enforcing an evidence-outranks-headcount principle, grading each specialist finding by its walkable failure path to resolve the noise and produce one accountable verdict.

How do I conduct a security-sensitive code review for a multi-lane diff?

Conducting a security-sensitive code review for a multi-lane diff involves convening the right specialist lanes, grading findings based on evidence, and converging the results into a finalized CodeReviewDecision artifact for the pipeline.

When do I need an evidence-based code review process instead of a standard review?

You need an evidence-based code review process when specialist findings require converging into one verdict, reviewers disagree about a diff, or a serious go/no-go review is required before verification runs or a change merges.

How are code review findings graded when reviewers disagree on a diff?

Code review findings are graded by the walkable failure path they expose, enforcing an evidence-outranks-headcount principle to ensure decisions are based on proven failure paths rather than reviewer count or confidence.

What does a code review lead do with specialist findings in a change pipeline?

A code review lead converges specialist findings into a single, evidence-based CodeReviewDecision artifact, recording failure paths and rationale to provide a finalized go/no-go verdict for the change pipeline.