dev-refactor

Analyze DONE features in feature.json and produce a consolidated refactor plan.

Updated Jan 1, 2026
One-click install
npx skills add https://github.com/AirMile/claude-config --skill dev-refactor-airmile
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dev-refactor
Source: https://github.com/AirMile/claude-config/tree/main/skills/dev-refactor
Command: npx skills add https://github.com/AirMile/claude-config --skill dev-refactor-airmile

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Batch refactor DONE features to improve code structure, naming, and patterns after feature completion, reducing technical debt and improving maintainability.

Core Features & Use Cases

  • Batch-first refactor workflow that analyzes all DONE features in parallel, triages CLEAN vs HAS_FINDINGS, and produces a single consolidated plan.
  • Creates one approved plan for applying changes with per-feature rollback and a unified testing phase.
  • Uses Context7 for stack-aware refactor patterns and enriches feature.json with status updates.

Quick Start

Run /dev-refactor when features are DONE to start batch refactoring across the project.

Frequently Asked Questions about dev-refactor

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

FAQPage Schema
How do I batch refactor completed features to reduce technical debt?▼

Batch refactor completed features by analyzing all pipeline files defined in feature.json in parallel. The tool triages refactor opportunities, enforces scope rules, and produces a single-approval plan to apply structural changes and run tests.

What is the best way to plan code refactoring across a project backlog?▼

Plan code refactoring across a project backlog using multi-phase analysis to triage DONE features into CLEAN or HAS_FINDINGS categories. This approach generates one consolidated plan with per-feature rollback and a unified testing phase.

How does automated refactor planning handle rollback and testing?▼

Automated refactor planning handles rollback and testing by creating a single approved plan that supports per-feature rollback, followed by a unified testing phase to validate all applied structural and naming improvements.

When should I run an automated code structure refactor on my pipeline?▼

Run an automated code structure refactor when features reach DONE status. This timing allows the batch analysis to improve naming and patterns across the backlog without disrupting active development.

Can I enforce scope rules during a batch code quality refactor?▼

Yes, you can enforce scope rules during a batch code quality refactor. The multi-phase analysis strictly enforces scope boundaries to ensure changes remain controlled and produce a single consolidated plan.

Does the refactor pipeline use stack-aware patterns for code improvements?▼

The refactor pipeline uses Context7 for stack-aware refactor patterns to guide code structure and naming improvements. It enriches feature.json with status updates throughout the batch processing workflow.