Can it handle very large pull requests?

To maintain high accuracy and avoid context window limitations, the skill automatically flags and skips pull requests containing more than 1,000 lines of changes.

What are CLAUDE.md files in this context?

CLAUDE.md files are project-specific guides. This skill reads them to ensure that the code changes align with your team's specific implementation patterns and best practices.

Does this skill run my build or unit tests?

No, it focuses on logic, context, and project guidelines. It is designed to complement your CI/CD pipeline, assuming that linters and test runners will handle syntax and build-related checks.

Does it provide line-specific feedback?

Yes, the skill generates formatted comments that include specific file paths, line numbers, and direct GitHub links with git SHAs for precise context.

How does the skill ensure high-quality feedback?

The skill uses five independent agents to audit the code from different perspectives and employs a scoring system that filters out any issues with a confidence score below 80%.

Professional Code Reviewer

Name: Professional Code Reviewer
Author: TheBushidoCollective

byTheBushidoCollective

•

协作工具

Automates multi-agent, high-confidence pull request reviews by analyzing code changes against project guidelines, historical context, and logic patterns.

关于

The Code Reviewer skill transforms Claude into a sophisticated pull request auditor using a multi-agent orchestration pattern. It coordinates five parallel Sonnet agents to evaluate PRs for logic bugs, adherence to local CLAUDE.md guidelines, historical git context, previous PR discussions, and inline comment compliance. By utilizing a rigorous deduplication and confidence-scoring system (0-100), the skill filters out noise and nitpicks, ensuring only verified, high-impact issues are reported back to GitHub. This is ideal for teams looking to maintain high code standards and reduce the manual burden on senior engineers during the review process.

主要功能

Confidence-based scoring system to eliminate false positives and nitpicks
Automated adherence checking for project-specific CLAUDE.md guidelines
Proof-of-work verification to ensure thorough agent analysis
Parallel multi-agent audits for bugs, style, and historical context
72 GitHub stars
Integration with GitHub CLI for automated, formatted PR commenting

使用场景

Enforcing consistent project standards across distributed development teams
Auditing complex changes against historical git context and previous PR feedback
Pre-screening pull requests to catch logic errors before human review

关于

主要功能

Confidence-based scoring system to eliminate false positives and nitpicks
Automated adherence checking for project-specific CLAUDE.md guidelines
Proof-of-work verification to ensure thorough agent analysis
Parallel multi-agent audits for bugs, style, and historical context
72 GitHub stars
Integration with GitHub CLI for automated, formatted PR commenting

使用场景

Enforcing consistent project standards across distributed development teams
Auditing complex changes against historical git context and previous PR feedback
Pre-screening pull requests to catch logic errors before human review