competitive-review

Dispatch two competing reviewers to analyze codebase tasks and merge findings.

5|1|Updated Nov 21, 2025

Search Tags:#architecture #security #quality #parallel #review #epistemic

Ready to run

npx skills add https://github.com/ANcpLua/ancplua-claude-plugins --skill competitive-review

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: competitive-review
Source: https://github.com/ANcpLua/ancplua-claude-plugins/tree/main/plugins/metacognitive-guard/skills/competitive-review
Command: npx skills add https://github.com/ANcpLua/ancplua-claude-plugins --skill competitive-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Running two reviewers in competition increases coverage and reduces blind spots in architectural and implementation analyses.

Core Features & Use Cases

Dual-reviewer workflow: arch-reviewer and impl-reviewer analyze independently.
Parallel execution: Faster results via concurrency.
Merged verdict: Combine results and highlight the most significant issues.

Quick Start

Before tackling a complex change, spawn both reviewers and compare findings.

Frequently Asked Questions about competitive-review

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

FAQPage Schema

How do I catch architectural issues that a single reviewer might miss?▼

Run two independent reviewers in parallel on your codebase task—an architecture reviewer and an implementation reviewer—to surface blind spots. Competing analyses increase coverage of security, quality, and design problems that single-pass review often overlooks.

When should I use parallel code review for complex technical decisions?▼

Use parallel review when tackling new code, modifying architecture, or making significant technical decisions. Multiple independent reviewers catch edge cases and security gaps faster than sequential review, especially for epistemic uncertainty in design choices.

How do I merge findings from multiple reviewers into a single verdict?▼

Dispatch both reviewers concurrently, then deduplicate and rank their results to highlight the most significant issues. The merged output feeds into deep-think-partner analysis to prioritize actionable quality and security improvements.

Can I run architectural and implementation reviews at the same time?▼

Yes. Parallel execution dispatches both the architecture reviewer and implementation reviewer simultaneously, reducing turnaround time while maintaining thoroughness across design and code-quality dimensions.

What's the difference between this approach and single-pass code review?▼

Competing reviewers reduce reviewer blind spots and epistemic gaps by analyzing the same task from independent angles. The merged verdict surfaces issues a single reviewer would miss, improving both security and architectural quality.