logan-roy

Orchestrate parallel code reviews of a branch diff and consolidate attributed findings.

Updated May 11, 2026
One-click install
npx skills add https://github.com/ruimfernandes/claude --skill logan-roy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: logan-roy
Source: https://github.com/ruimfernandes/claude/tree/main/skills/logan-roy
Command: npx skills add https://github.com/ruimfernandes/claude --skill logan-roy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It resolves the time-consuming back-and-forth of reviewing a branch by running multiple independent code reviewers, consolidating their findings with attribution, and applying only the changes you explicitly approve.

Core Features & Use Cases

  • Parallel multi-reviewer orchestration: Executes Mr. Milchick, the CodeRabbit CLI review, and the sona-marketplace code-reviewer against the current branch diff (master..HEAD).
  • Consolidated, deduped recommendations with attribution: Merges overlapping findings across reviewers, groups them into consensus vs single-reviewer actionable items, and preserves which reviewer flagged each change.
  • Guardrails and safe apply: Enforces a clean working tree gate, requires explicit approval before editing code, applies minimal edits, and retries review continuity even if a reviewer lane fails.

Quick Start

Ask: "Logan Roy, review this branch and clean it up, then propose the approved changes after you summarize what will be applied."

Frequently Asked Questions about logan-roy

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

FAQPage Schema
How do I run multiple code reviewers in parallel on a git diff?

To run parallel code review on a git diff, this Skill orchestrates multiple independent reviewers against your current branch diff (master..HEAD), consolidating findings into a deduped list with attribution. It retries failed reviewer lanes to ensure continuity.

What is a consensus code review and how does deduplication work?

Consensus code review merges overlapping findings from multiple independent reviewers, grouping them into consensus versus single-reviewer actionable items. This deduplication preserves reviewer attribution so you know exactly which reviewer flagged each recommended change.

Do I need a clean git working tree for automated branch cleanup?

Yes, a clean git working tree is a strict prerequisite for this automated branch cleanup workflow. The Skill enforces this guardrail before collecting per-lane reports, requiring explicit approval before applying any minimal file edits to your code.

How to apply only explicitly approved fixes from a code review?

To apply approved fixes safely, the Skill generates an approval plan from consolidated reviewer findings and applies only the changes you explicitly approve. It makes minimal file edits while preserving reviewer attribution for every applied fix.