request-refactor-plan
๐ฏSkillfrom mattpocock/skills
|A skill that interviews the developer about a refactoring problem, explores the codebase to verify assertions, evaluates alternative approaches, checks test coverage, and creates a detailed GitHub issue with a plan of tiny, incremental commits.
Overview
Request Refactor Plan is a skill from Matt Pocock's "Skills For Real Engineers" collection that turns a vague refactoring idea into a well-scoped, actionable plan. It interviews you about the refactoring problem, explores the codebase to verify your assertions, evaluates alternative approaches, checks existing test coverage, and produces a detailed GitHub issue ready for implementation.
Key Features
- Interactive grilling session - Asks targeted questions to understand the refactoring motivation, scope, and constraints before jumping into a plan
- Codebase-aware verification - Explores the actual code to validate assumptions, find affected areas, and surface risks you might have missed
- Alternative approach evaluation - Considers multiple ways to achieve the refactoring goal and documents trade-offs between them
- Test coverage analysis - Checks what tests exist for the code being refactored so nothing breaks silently during the change
- Ready-to-implement output - Produces a structured GitHub issue with clear acceptance criteria rather than a vague description
Who is this for?
- Engineers planning non-trivial refactors who want a structured process to reduce risk and avoid scope creep
- Tech leads who need to document and communicate refactoring plans to their team before starting work
- Developers using Claude Code or similar agents who want to avoid the common failure of jumping straight into refactoring without a plan
Same repository
mattpocock/skills(55 items)
Installation
npx vibeindex add mattpocock/skills --skill request-refactor-plannpx skills add mattpocock/skills --skill request-refactor-plan~/.claude/skills/request-refactor-plan/SKILL.mdSKILL.md
More from this repository10
Interviews the user relentlessly about every aspect of a plan or design until reaching shared understanding, resolving each branch of the decision tree one by one. Provides recommended answers for each question and explores the codebase when answers can be found there.
A skill that runs a grilling session to challenge your plan against an existing domain model, sharpen terminology by building a shared language in CONTEXT.md, and document architectural decisions in ADRs before coding begins.
An agent skill from Matt Pocock's engineering toolkit that helps improve your codebase architecture. Part of a composable skill set designed to fix common AI coding agent failure modes through structured, experience-based workflows.
Enforces test-driven development with a strict red-green-refactor loop using vertical slices (one test then one implementation at a time), emphasizing behavior verification through public interfaces rather than implementation-coupled testing.
Initial setup skill for Matt Pocock's engineering skills collection that configures your backlog manager, triage labels, and documentation paths for AI coding agents.
A collection of small, composable engineering skills designed for real development workflows, featuring grilling sessions for requirement alignment, issue tracking integration, and task triage with customizable labels.
A composable skill for triaging issues and tickets using configurable labels, designed to fix common failure modes in AI coding agents by providing structured backlog management workflows.
A skill that turns the current conversation context and codebase understanding into a structured PRD (Product Requirements Document) and submits it as a GitHub issue, with modular architecture and user story generation.
Breaks a plan, spec, or PRD into independently-grabbable GitHub issues using tracer-bullet vertical slices. Each issue is a thin end-to-end path through every integration layer that can be demoed or verified on its own.
A collection of composable engineering skills by Matt Pocock designed to fix common AI coding agent failure modes: misalignment via grilling sessions, verbosity through shared domain language (CONTEXT.md), code quality with TDD red-green-refactor loops, and codebase entropy with architecture improvement workflows.