Code development and quality assurance tools for systematic analysis, testing, debugging, and review. The Development plugin provides essential tools for building high-quality code. It includes commands for codebase analysis, test-driven development, debugging, code execution, and comprehensive code review. Three specialized agents provide expert assistance for architecture, testing, and code qual
Add this skill
npx mdskills install applied-artificial-intelligence/developmentComprehensive development toolkit with well-structured commands, excellent documentation, and clear usage examples
Code development and quality assurance tools for systematic analysis, testing, debugging, and review.
The Development plugin provides essential tools for building high-quality code. It includes commands for codebase analysis, test-driven development, debugging, code execution, and comprehensive code review. Three specialized agents provide expert assistance for architecture, testing, and code quality.
/analyze [focus_area] [--with-thinking] [--semantic]Analyze ANY project to understand its structure and architecture with semantic code intelligence and structured reasoning.
What it does:
Usage:
/analyze # Full project analysis
/analyze src/auth # Analyze specific area
/analyze --with-thinking # Use structured reasoning
/analyze --semantic # Use Serena semantic analysis
/analyze src/api --with-thinking --semantic # Deep analysis with all tools
Analysis Types:
Output:
When to use:
/test [tdd] [pattern]Test-driven development workflow using test-engineer agent. Create tests, analyze coverage, and ensure quality.
What it does:
Usage:
/test # Run existing tests
/test tdd # Start TDD workflow
/test src/auth/* # Test specific pattern
/test --coverage # Generate coverage report
TDD Workflow:
Test Types Supported:
When to use:
/fix [error|review|audit|all] [file/pattern]Universal debugging and fix application with semantic code analysis. Debug errors or apply review fixes automatically.
What it does:
Usage:
/fix # Fix latest error
/fix error src/auth/login.js # Fix specific file errors
/fix review # Apply review suggestions
/fix audit # Fix audit findings
/fix all # Fix everything
Fix Sources:
Fix Process:
When to use:
/review [file/directory] [--spec requirements.md] [--systematic] [--semantic]Standard code review focused on bugs, design flaws, dead code, and code quality with prioritized action plan.
What it does:
Usage:
/review # Review recent changes
/review src/auth # Review directory
/review src/auth/login.js # Review specific file
/review --spec @requirements.md # Review against spec
/review --systematic # Use structured reasoning
/review --semantic # Use semantic code analysis
Review Focus Areas:
Output Format:
HIGH Priority:
- [BUG] Null pointer in login.js:45
- [SECURITY] SQL injection risk in query.js:120
MEDIUM Priority:
- [DESIGN] God class in UserService.js
- [PERFORMANCE] N+1 query in getUsers()
LOW Priority:
- [QUALITY] Inconsistent naming in helpers.js
- [DEAD CODE] Unused import in utils.js
When to use:
/prepare-review [focus_area_or_context]Prepare external code review package using RepoMix with intelligent file selection and token-efficient formatting.
What it does:
Usage:
/prepare-review # Generic quality review
/prepare-review "API authentication layer" # Focus on specific area
/prepare-review "Data pipeline error handling" # Focus on specific concerns
Review Package Includes:
Output Location:
.claude/external_reviews/YYYY-MM-DD-HHMM/
└── review_package.md # Complete package ready to share
Token Efficiency:
File Selection Intelligence:
Review Focus Areas (default):
When to use:
Requirements:
npm install -g repomix) or available via npxExample Output:
✅ External review package prepared!
Location: .claude/external_reviews/2025-11-14-0530/review_package.md
Package Summary:
- Focus: API authentication layer
- Files included: 12 files (core auth modules)
- Token count: ~47,000 tokens ✅
- Format: XML (no line numbers)
Next Steps:
1. Review the package
2. Copy entire review_package.md
3. Submit to external reviewer
architect.md)System design and architectural decisions specialist. Provides high-level design guidance.
Expertise:
Capabilities:
When to use:
/agent architect "Design authentication system for multi-tenant SaaS"
/agent architect "Evaluate microservices vs monolith for our use case"
/agent architect "Review proposed API architecture"
test-engineer.md)Test creation, coverage analysis, and quality assurance specialist.
Expertise:
Capabilities:
When to use:
/agent test-engineer "Create comprehensive tests for auth module"
/agent test-engineer "Analyze test coverage and suggest improvements"
/agent test-engineer "Design testing strategy for new feature"
code-reviewer.md)Code review, documentation quality, and security audit specialist.
Expertise:
Capabilities:
When to use:
/agent code-reviewer "Review authentication implementation for security"
/agent code-reviewer "Assess code quality of new payment module"
/agent code-reviewer "Review PR #123 for merge readiness"
/agent for invoking development agents/system:status to show analysis and review progress/performance for execution metrics/analyze used during /explore phase/test and /review used during /ship validation/fix helps resolve issues blocking /next/review provides feedback before git commit/fix applies changes that git commit includes.claude/config.json){
"development": {
"analyze": {
"defaultMode": "semantic",
"useStructuredReasoning": true,
"depthLevel": "comprehensive"
}
}
}
{
"development": {
"test": {
"framework": "auto-detect",
"coverageThreshold": 80,
"runBeforeCommit": true
}
}
}
{
"development": {
"review": {
"autoReview": false,
"focusAreas": ["bugs", "security", "design"],
"severityThreshold": "medium"
}
}
}
Graceful Degradation: All commands work without MCP tools.
# 1. Start with high-level analysis
/analyze
# 2. Deep-dive into specific areas
/analyze src/auth --semantic --with-thinking
# 3. Review findings and plan improvements
# 1. Start TDD mode
/test tdd
# 2. Write test first
# (test-engineer helps generate test cases)
# 3. Run test (should fail)
npm test
# 4. Implement feature
# 5. Run test (should pass)
npm test
# 6. Refactor
# 7. Repeat
# 1. Self-review before PR
/review --semantic
# 2. Fix HIGH priority issues
/fix review
# 3. Address MEDIUM priority
# (make changes manually or with /fix)
# 4. Re-review to verify
/review
# 5. Submit PR when clean
When Serena MCP is available:
Enable Serena:
Serena is enabled via MCP configuration (not a wrapper command). See docs/mcp-setup.md for setup instructions.
When Sequential Thinking is available:
Automatic: Used automatically by architect and code-reviewer when available
--with-thinking for deeper analysis--semantic if Serena available/analyze src/problem-area/fix error [file]/fix review after running /review/review src/moduleThe development plugin tracks:
View with:
/system:status verbose
/performance
# Start TDD workflow
/test tdd
# Test-engineer generates test cases
# Implement feature to pass tests
# Run tests
npm test
# Review implementation
/review src/new-feature
# Fix any issues
/fix review
# Ship it
/ship --commit
# Analyze problem area
/analyze src/buggy-module --semantic
# Review code for bugs
/review src/buggy-module
# Fix identified issues
/fix review
# Run tests
npm test
# Verify fix
/test src/buggy-module
# Get architect input
/agent architect "Review proposed microservices architecture"
# Analyze current architecture
/analyze --with-thinking
# Identify issues
/review --systematic
# Plan improvements based on findings
MIT License - see LICENSE for details.
Version: 1.0.0 Category: Development Commands: 7 (analyze, docs, fix, git, prepare-review, review, test) Agents: 3 (architect, test-engineer, code-reviewer) Dependencies: core (^1.0.0) MCP Tools: Optional (sequential-thinking, serena, context7)
Best experience: Claude Code
/plugin marketplace add applied-artificial-intelligence/developmentThen /plugin menu → select skill → restart. Use /skill-name:init for first-time setup.
Other platforms
Install via CLI
npx mdskills install applied-artificial-intelligence/developmentDevelopment Plugin is a free, open-source AI agent skill. Code development and quality assurance tools for systematic analysis, testing, debugging, and review. The Development plugin provides essential tools for building high-quality code. It includes commands for codebase analysis, test-driven development, debugging, code execution, and comprehensive code review. Three specialized agents provide expert assistance for architecture, testing, and code qual
Install Development Plugin with a single command:
npx mdskills install applied-artificial-intelligence/developmentThis downloads the skill files into your project and your AI agent picks them up automatically.
Development Plugin works with Claude Code, Claude Desktop, Cursor, Vscode Copilot, Windsurf, Continue Dev, Codex, Gemini Cli, Amp, Roo Code, Goose, Opencode, Trae, Qodo, Command Code. Skills use the open SKILL.md format which is compatible with any AI coding agent that reads markdown instructions.