improve

Orchestrate iterative code quality improvement cycles with adversarial discussions.

Updated Jul 5, 2025
One-click install
npx skills add https://github.com/sqfzy/my_configs --skill improve-sqfzy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: improve
Source: https://github.com/sqfzy/my_configs/tree/main/claude/skills/improve
Command: npx skills add https://github.com/sqfzy/my_configs --skill improve-sqfzy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Structured code quality improvement cycle helps teams identify latent issues, discuss them adversarially, and capture lessons learned, turning maintenance into a disciplined, measurable process.

Core Features & Use Cases

  • Global assessment and architecture mapping to surface design flaws and opportunities.
  • Iterative improvement with multi-role discussion, implementation, verification, and documentation.
  • Auto-saves improvement reports to .artifacts/ upon trigger events like debt cleanup or code health checks.

Quick Start

Specify a target file or module to start the improvement cycle and run the iterative reviews until convergence.

Frequently Asked Questions about improve

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

FAQPage Schema
How do I perform a structured code quality improvement cycle for an ongoing software project?

A structured code quality improvement cycle identifies latent issues through global assessment, discusses them adversarially, and captures lessons learned. This turns ongoing maintenance into a disciplined, measurable process across your modules.

What is the best way to clean up technical debt and run periodic code health checks?

The best way to clean up technical debt is using an iterative improvement process with multi-role discussion, implementation, and verification. It automatically saves code health check reports to .artifacts/ upon trigger events.

How does an adversarial discussion loop work for architecture mapping and code review?

An adversarial discussion loop for code review surfaces design flaws by forcing multi-role debate over identified issues. It maps global architecture to ensure iterative reviews converge toward measurable software improvement.

Can I specify target files and depth for an iterative code review and improvement process?

Yes, you can specify target files, depth, and auto mode to start the improvement cycle. This allows you to run iterative reviews on specific modules until the code quality convergence is reached.

When do I need a structured code maintenance process versus ad-hoc refactoring?

You need a structured code maintenance process when managing ongoing tech debt cleanup across large modules. Unlike ad-hoc refactoring, it provides iterative verification, multi-role discussion, and auto-saved improvement reports.