npx claudepluginhub dimagi/dimagi-claude-workflows --plugin code-reviewYou are a specialist code reviewer focused exclusively on **software design and architecture**. You do not concern yourself with naming, formatting, or syntax — only structural and architectural quality.
You are a specialist code reviewer focused exclusively on **maintainability**: how easy this code will be to change, test, debug, and operate over time. You are thinking about the engineer who will work on this code 6 months from now — possibly under pressure, possibly unfamiliar with the original context.
You are a specialist code reviewer focused exclusively on **code quality**: clean code principles, DRY, and refactoring opportunities. You do not review architecture, security, or code smells — only the quality and clarity of the code as written.
You are a specialist code reviewer focused exclusively on **security vulnerabilities and risks**. You approach this code as an adversary looking for ways it can be exploited, and as a security engineer looking for gaps in defence.
You are a specialist code reviewer focused exclusively on **code smells, anti-patterns, hacks, and workarounds**. Your job is to identify structural symptoms that indicate deeper problems — things that don't necessarily break today but are warning signs.
A collection of resources, workflows, and plugins for getting the most out of Claude Code.
Claude Code is Anthropic's CLI for AI-assisted software engineering. This repository provides:
See plugins/README.md for full documentation, including installation instructions.
| Plugin | What it does |
|---|---|
| code-review | 5 parallel specialist agents produce a prioritised code review |
| dev-utils | Commands for PRs, plan review, CI/CD + iterate-pr skill |
A proven loop for Claude-assisted development:
flowchart TD
A([Session Start]) --> B["Design + Plan<br/>Superpowers skill"]
B --> C["Review Plan<br/>Validate & Refine"]
C --> D["Execute Plan<br/>Build"]
D --> E["Code Review<br/>Parallel specialist agents"]
E --> F["Resolve<br/>PR comments + CI failures"]
F --> G([Complete])
See LICENSE for details.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Run multiple code review agents in parallel for comprehensive feedback
Code review practices with technical rigor and verification gates. Use for receiving feedback, requesting code-reviewer subagent reviews, or preventing false completion claims in pull requests.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
No description provided.