A 7-agent refactoring team that explores, researches, tests, plans, challenges, executes, and verifies code improvements. Runs autonomously with gated decisions at planning and verification stages. Includes comprehensive skills for code quality, architecture, design, security, and documentation.
Quality gate before execution. Reviews the refactoring roadmap for feasibility (can it be executed?) and semantic correctness (will it improve clarity?). Produces detailed feedback per slice. Routes back to Planner for fixable issues, blocks only for critical problems. Hands off approved roadmap to Refactorer.
Entry point for the refactoring team. Deep dives into any codebase to understand what exists: architecture, modules, dependencies, patterns, how pieces fit together. Documents the current state comprehensively. Makes light notes on potential improvement areas (naming, organization, documentation gaps) without deep analysis. Hands off complete project understanding to Researcher.
Creates comprehensive, prioritized refactoring roadmap. Takes Explorer's understanding, Researcher's best practices, and Tester's coverage assessment. Builds detailed roadmap organized into phases (Small/Medium/Large) with specific, actionable slices. Sequences with dependency awareness. Balances impact, risk, and feasibility. Enforces git as mandatory. Hands off to Challenger for review.
Executes the approved refactoring roadmap with discipline. Follows slices in order, commits per strategy, tests after each slice, maintains clean git history. Autonomously fixes issues during execution. Only starts with Challenger-approved roadmap. Produces practical summary of what changed. Hands off to Verifier.
Takes Explorer's project understanding and identifies best practices for this specific codebase type. Researches universal refactoring principles AND project-type-specific standards (React, Node, Python, CLI, etc.). Weights project-type conventions heavier. Documents findings and makes specific recommendations. Hands off to Tester.
Architecture principles for module boundaries, design patterns, and migration strategies
Non-negotiable code quality standards for structure, testing, naming, and documentation
Baseline coding conventions for naming, formatting, and style
Design and UI standards for accessibility, semantic HTML, and responsive layouts
Documentation standards for changelogs, feature specs, and module documentation
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Personal Claude Code customizations: multi-agent teams, standalone commands, and agents.
claude-customizations/
├── teams/ # Multi-agent plugin teams (installed via marketplace)
├── commands/ # Standalone command docs (reference only)
└── agents/ # Standalone agent docs (reference only)
Installable plugin teams with multiple coordinated agents.
| Team | Description | Version |
|---|---|---|
| refactor-team | 7-agent refactoring workflow | 2.0.0 |
| implement-team | 5-agent TDD implementation | 1.0.0 |
| diagnose-team | 5-agent diagnostic workflow | 1.0.0 |
# Add marketplace (one-time)
claude plugin marketplace add https://github.com/AlexanderStephenThompson/claude-customizations
# Install teams
claude plugin install refactor-team
claude plugin install implement-team
claude plugin install diagnose-team
| Command | Description |
|---|---|
/refactor-team:refactor [path] [focus] | Run 7-agent refactoring workflow |
/implement-team:implement <feature> | Run 5-agent TDD implementation workflow |
/diagnose-team:diagnose <problem> | Run 5-agent diagnostic workflow |
Standalone commands stored in ~/.claude/commands/ (reference docs only).
| Command | Description |
|---|---|
| /commit | Stage all changes and create conventional commit |
Standalone agents stored in ~/.claude/agents/ (reference docs only).
| Agent | Description |
|---|---|
| Codebase Scout | Explore and document unfamiliar codebases |
| Improvement Auditor | Deep-dive codebase analysis for refactoring |
See SYNC-GUIDE.md for instructions on syncing this repository with local templates.
~/.claude/npx claudepluginhub AlexanderStephenThompson/claude-customizations --plugin refactor-teamSlash commands that launch Agent Teams for research, debugging, and feature implementation with code review gates
Interactive skill that analyzes a task, proposes an agent team composition, and creates the team after user confirmation
AI team workflow harness for Claude Code — specialist agents, shared skills, slash commands, hooks, and staged delegation pipeline.
Unified review, refactoring, and cleanup workflows with specialized analysis agents
Refactor code following best practices and design patterns
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.