Custom Review Agents

Custom AI agents for code review and test generation with domain-specific skills.

Overview

This plugin provides 4 specialized agents and 12 domain knowledge skills following the Every marketplace methodology:

• Agents = Personalities and perspectives (how to review)
• Skills = Domain knowledge (what to know)

Architecture

agents/              # 4 review/generation agents
├── review/
│   ├── go-reviewer.md
│   ├── graphql-reviewer.md
│   └── nextjs-reviewer.md
└── workflow/
    └── playwright-generator.md

skills/              # 12 domain knowledge skills
├── go-concurrency/       # Goroutines, channels, context
├── go-performance/       # Memory, strings.Builder, I/O
├── go-testing/           # Table-driven tests, race detector
├── go-error-handling/    # Error wrapping, errors.Is/As
├── graphql-schema-design/    # Types, pagination, nullability
├── graphql-resolvers/        # Dataloader, N+1, auth
├── graphql-mutations/        # Payloads, field errors
├── nextjs-server-components/ # Server/Client, data fetching
├── nextjs-performance/       # Images, fonts, caching
├── nextjs-metadata/          # SEO, OpenGraph
├── playwright-page-objects/  # Page Object Model
└── playwright-best-practices/ # Selectors, waits, a11y

Agents

go-reviewer

Reviews Go code for idiomatic patterns, concurrency safety, and performance.

Skills used: go-concurrency, go-performance, go-testing, go-error-handling

graphql-reviewer

Reviews GraphQL schemas, resolvers, and mutations for best practices and performance.

Skills used: graphql-schema-design, graphql-resolvers, graphql-mutations

nextjs-reviewer

Reviews Next.js code for framework best practices and performance.

Skills used: nextjs-server-components, nextjs-performance, nextjs-metadata

playwright-generator

Generates E2E tests with Page Object Model and best practices.

Skills used: playwright-page-objects, playwright-best-practices

Installation

1. Copy this directory to your Claude Code plugins folder
2. Restart Claude Code
3. Agents and skills will be available

Usage

Agents are automatically invoked based on context, or explicitly:

"Have the Go reviewer review this code"
"Review this GraphQL schema for N+1 queries"
"Check this Next.js component for proper Server Component usage"
"Generate Playwright tests for the checkout flow"

Skills are automatically loaded by agents when needed.

License

MIT