project-starter
A universal Claude Code workflow plugin with specialized agents, skills, hooks, and output styles for any software project.
Table of Contents
- Installation
- Features
- Usage Examples
- Configuration
- Extending the Plugin
- Plugin Structure
- Requirements
- Contributing
- Author
- License
Installation
# Clone the plugin
git clone https://github.com/CloudAI-X/claude-workflow.git
# Run Claude Code with the plugin
claude --plugin-dir ./claude-workflow
Features
7 Specialized Agents
| Agent | Purpose | Auto-Trigger Keywords |
|---|---|---|
orchestrator | Coordinate complex multi-step tasks | "improve", "refactor", multi-module changes |
code-reviewer | Review code quality and best practices | After code changes, before commits |
debugger | Systematic bug investigation and fixing | Errors, test failures, crashes |
docs-writer | Create technical documentation | README, API docs, guides |
security-auditor | Security vulnerability detection | Auth, user input, sensitive data |
refactorer | Code structure improvements | Technical debt, cleanup |
test-architect | Design comprehensive test strategies | Adding/improving tests |
6 Knowledge Skills
- project-analysis - Understand any codebase structure and patterns
- testing-strategy - Design test approaches (unit, integration, E2E)
- architecture-patterns - System design guidance (Clean Architecture, Hexagonal, etc.)
- performance-optimization - Speed up applications, identify bottlenecks
- git-workflow - Version control best practices, conventional commits
- api-design - REST/GraphQL API patterns and best practices
4 Output Styles (Slash Commands)
Switch work modes with namespaced slash commands:
/project-starter:architect # System design mode - focus on architecture before code
/project-starter:rapid # Fast development mode - ship quickly, iterate
/project-starter:mentor # Learning/teaching mode - explain the "why"
/project-starter:review # Code review mode - strict quality standards
8 Automation Hooks
| Hook | Trigger | Action |
|---|---|---|
| Security scan | Edit/Write | Blocks commits with potential secrets |
| File protection | Edit/Write | Blocks edits to lock files, .env, .git |
| Auto-format | Edit/Write | Runs prettier/black/gofmt based on file type |
| Command logging | Bash | Logs commands to .claude/command-history.log |
| Environment check | Session start | Validates Node.js, Python, Git availability |
| Prompt analysis | User prompt | Suggests appropriate agents |
| Input notification | Input needed | Desktop notification when Claude needs input |
| Complete notification | Task complete | Desktop notification when task finishes |
Configuration
Permissions
See PERMISSIONS.md for recommended permissions to add to your project's .claude/settings.local.json.
MCP Servers
See mcp-servers-template.md for common MCP server configurations (GitHub, Sentry, databases, etc.).
Usage Examples
Starting a New Project
User: What's the architecture of this project?
Claude: [Uses project-analysis skill to explore and explain the codebase]
Implementing Features
User: Add user authentication
Claude: [Uses orchestrator agent for multi-step planning and delegation]
Fixing Bugs
User: The login is broken
Claude: [Uses debugger agent for systematic root cause analysis]
Code Review
User: Review my changes
Claude: [Uses code-reviewer agent with correctness, security, performance checks]
Switching Modes
User: /project-starter:architect
Claude: [Switches to architecture mode - designs before implementing]