Why does the skill avoid using the TaskOutput tool?

It uses file-based reports to maintain context efficiency. Large tool outputs can clutter the chat history and degrade the model's performance on long-running tasks.

Does this skill automatically modify my source code?

No, the evaluation agents are read-only. They report findings to markdown files, and you decide which improvements to apply during the iterative fixing phase.

When is the documentation review performed?

The documenter agent runs last, after all code fixes have been applied. This ensures the documentation is evaluated against the final, corrected state of the code.

How many agents are included in this skill?

This skill orchestrates eight specialized agents: namer, code-simplifier, comment-cleaner, test-reviewer, error-handling-reviewer, security-reviewer, observability-reviewer, and a documenter.

Code Quality Evaluation (The Crucible)

Name: Code Quality Evaluation (The Crucible)
Author: xobotyi

byxobotyi

•

安全与测试

Orchestrates eight specialized AI agents to perform deep code reviews across naming, security, complexity, and testing dimensions.

The Code Quality Evaluation skill transforms Claude into a comprehensive review board by launching eight specialized background agents to analyze your codebase simultaneously. It systematically evaluates naming conventions, architectural complexity, test coverage, security vulnerabilities, and observability without modifying your source code during the audit phase. By producing localized markdown reports instead of consuming chat context with tool outputs, it maintains high token efficiency while providing granular, actionable insights that you can selectively apply to ensure your implementation meets production-grade standards.

主要功能

01Orchestrates 7 parallel background agents for rapid multi-dimensional analysis

02Analyzes naming, complexity, tests, security, error handling, and observability

03Uses file-based reporting to preserve context efficiency and model focus

04Automates the setup and cleanup of temporary review artifacts

051 GitHub stars

06Includes a specialized documentation agent that runs after code fixes

使用场景

01Automated pull request reviews for architecture and security compliance

02Refactoring legacy codebases to improve maintainability and test coverage

03Performing comprehensive pre-commit audits on complex features

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add xobotyi/cc-foundry code-quality-evaluation

For use in Claude.ai and ChatGPT

Download Skill