How does it handle project-specific documentation?

The skill checks for plan documents in docs/plans/ to ensure that code changes align with the original acceptance criteria and project requirements.

How does the multi-agent system work?

It spawns three distinct Claude sub-tasks focused specifically on security, performance, and architecture. This ensures specialized focus for each domain before synthesizing results into a single report.

Can it fix the issues it finds?

Yes, in interactive mode you can choose to fix specific findings or all high-priority items, and the skill will automatically apply remediations following your project's patterns.

Does it report on linting or style issues?

No, this skill is designed to ignore formatting and style issues, focusing instead on logic, security vulnerabilities, and architectural improvements.

Can I limit the scope of the review?

Yes, you can provide specific file paths, a Pull Request number, or a custom scope. It also automatically detects changed or staged files if no arguments are provided.

Expert Code Review Panel

Name: Expert Code Review Panel
Author: noodlemind

bynoodlemind

0•

セキュリティとテスト

Performs deep code analysis using parallel expert agents for security, performance, and architectural integrity.

This skill elevates the code review process by deploying a three-expert panel of specialized sub-agents that independently evaluate your codebase for security vulnerabilities, performance bottlenecks, and structural alignment. By synthesizing findings into a prioritized, deduplicated report, it provides actionable insights and automated fixes that respect your existing conventions while intentionally ignoring trivial style or formatting issues, allowing developers to focus on high-impact logic and quality.

主な機能

01Intelligent scope detection via git diffs and PR numbers

02Automated remediation of identified issues with code context awareness

03Automated synthesis and deduplication of findings into prioritized reports

04Plan compliance verification against local design documentation

05Parallel multi-agent analysis for security, performance, and architecture

060 GitHub stars

ユースケース

01Ensuring new implementation logic aligns with existing architectural standards and project plans

02Auditing a Pull Request for security vulnerabilities and performance regressions before merging

03Rapidly identifying and fixing critical technical debt across specific directories

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add noodlemind/nanodex-marketplace code-review

For use in Claude.ai and ChatGPT

Download Skill