Does it support security auditing specifically?

Yes, it includes a dedicated Security Auditor Agent that specifically checks for OWASP Top 10 vulnerabilities, secrets exposure, and dependency flaws.

What are the different review modes available?

It offers 'Quick' for fast automated feedback, 'Standard' for full parallel agent analysis, and 'Thorough' for deep architectural and type design reviews.

How does this differ from a standard AI code review?

Unlike single-prompt reviews, this skill uses specialized agents for logic, security, and architecture, providing significantly higher accuracy and depth on complex code changes.

Can I integrate this with my CI/CD pipeline?

Yes, the skill is designed to work with GitHub Actions and other CI tools to automate PR reviews and enforce pre-defined quality gates.

How does the consensus mechanism work?

It aggregates findings from multiple agents, deduplicates similar issues, and boosts severity when multiple specialized agents identify the same vulnerability or logic error.

Code Review Pipeline

Name: Code Review Pipeline
Author: pagerguild

bypagerguild

安全与测试

Orchestrates specialized AI agents to perform comprehensive code reviews, security audits, and architectural analysis using a scatter-gather consensus pattern.

关于

The Code Review Pipeline skill implements a sophisticated multi-agent orchestration workflow designed to outperform single-agent reviews by utilizing specialized personas for logic, security, architecture, and testing. It follows a structured three-stage process—automated checks, parallel agent reviews, and consensus aggregation—to provide high-confidence feedback on complex code changes. By deduplicating findings and prioritizing issues based on cross-validation, this skill ensures that developers receive actionable, prioritized reports that maintain high code quality and security standards without the noise of typical automated tools.

主要功能

Parallel execution of subagents to minimize feedback latency during PR reviews
0 GitHub stars
Three-tier review modes ranging from rapid WIP checks to thorough architectural audits
Automated integration with quality gates to block critical security vulnerabilities
Multi-agent orchestration with specialized logic, security, and architecture auditors
Consensus-driven reporting that deduplicates findings and validates severity

使用场景

Validation of complex AI-generated code blocks before merging into production
Enforcing team-wide quality standards and coding patterns in CI/CD pipelines
Comprehensive PR reviews requiring deep security and architectural analysis

关于

主要功能

Parallel execution of subagents to minimize feedback latency during PR reviews
0 GitHub stars
Three-tier review modes ranging from rapid WIP checks to thorough architectural audits
Automated integration with quality gates to block critical security vulnerabilities
Multi-agent orchestration with specialized logic, security, and architecture auditors
Consensus-driven reporting that deduplicates findings and validates severity

使用场景

Validation of complex AI-generated code blocks before merging into production
Enforcing team-wide quality standards and coding patterns in CI/CD pipelines
Comprehensive PR reviews requiring deep security and architectural analysis