looping-quality

Iteratively scan, fix, and re-scan code until quality findings reach zero.

125|42|Updated May 19, 2025
One-click install
npx skills add https://github.com/SocketDev/socket-mcp --skill looping-quality
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: looping-quality
Source: https://github.com/SocketDev/socket-mcp/tree/main/.claude/skills/fleet/looping-quality
Command: npx skills add https://github.com/SocketDev/socket-mcp --skill looping-quality

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates iterative quality scanning by looping a scan, applying fixes, and re-scanning to converge on a clean quality report in an interactive workflow that commits changes.

Core Features & Use Cases

  • Orchestrates iterative scans by invoking the scanning-quality skill, collecting findings, and tracking an iteration counter.
  • Applies fixes via the refactor-cleaner agent with CLAUDE.md pre-action protocol: dead code first, then structural changes, limited to a small number of files per phase.
  • Re-scan and verify builds and tests after each fix, stopping when zero findings remain or the iteration cap is reached.
  • Commits per iteration with a clear subject like "fix: iterate quality scan (N)" to maintain a visible history.

Quick Start

Start an iterative quality loop by invoking looping-quality to scan, fix, and re-scan until the findings are eliminated or the maximum iterations are reached.

Frequently Asked Questions about looping-quality

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

FAQPage Schema
How do I automate iterative code quality scans and auto-fixes?

Automating iterative code quality scans is done by looping a scanner, applying fixes via a refactor agent, and re-scanning until zero findings remain. This workflow commits changes per iteration and validates the build and tests after each fix.

What is the best way to progressively reduce codebase quality findings?

The best way to progressively reduce quality findings is enforcing an iterative scan-and-fix loop that applies fixes in phases—dead code first, then structural changes—limited to a small number of files per phase to maintain stability.

How does an iterative quality scan workflow handle commit tracking?

An iterative quality scan workflow handles commit tracking by committing changes per iteration with a clear subject like "fix: iterate quality scan (N)", maintaining a visible history of progressive codebase fixes.

Does the automated quality fix loop validate tests after each iteration?

Yes, the automated quality fix loop validates the build and tests after each iteration. This ensures that applied refactoring changes do not break existing functionality before proceeding to the next scan cycle.

What happens when an iterative code refactor loop reaches the iteration cap?

When an iterative code refactor loop reaches the iteration cap, it stops applying fixes. The workflow terminates either when zero quality findings remain or the maximum iteration limit is reached to prevent infinite loops.

Can I use an iterative scan and fix workflow for large codebases?

You can use an iterative scan and fix workflow for large codebases because it limits structural changes to a small number of files per phase. This phased approach prevents overwhelming the refactor agent during dead code removal and structural updates.