claude-setup 1.0.0

Intelligent scaffolding system for Claude Code configurations - Generate optimal .claude setups for any project

claude-setup

Intelligent scaffolding system for optimal Claude Code configurations.

Description

Generate optimal .claude configurations for any project by automatically detecting your tech stack and project structure. Creates customized settings, commands, agents, and workflows tailored to your specific development needs.

Installation

Using uv (Recommended)

uvx claude-setup init

Using pipx

pipx install claude-setup

Using pip

pip install claude-setup

Quick Start

Initialize Claude in your project:

cd your-project
claude-setup init

Configure automatically:

claude-setup configure --auto

Or interactively:

claude-setup configure --interactive

Usage

Core Commands

# Initialize new configuration
claude-setup init [--template NAME] [--force]

# Configure project
claude-setup configure [--auto] [--interactive]

# List templates
claude-setup template list

# Apply template
claude-setup template apply <name>

# Check status
claude-setup status [--check] [--validate]

# Export configuration
claude-setup export -o config.json

Self-Configuration

Start Claude and request:

"Analyze this project and configure yourself optimally for [your project description]"

Claude will generate the perfect configuration using the built-in scaffolding agent.

Configuration

Directory Structure

.claude/
├── settings.json      # Permissions and model config
├── CLAUDE.md         # Project context
├── commands/         # Custom slash commands
├── agents/           # Specialized AI agents
├── hooks/            # Event automation
└── templates/        # Code templates

settings.json

{
    "model": "claude-opus-4-1-20250805",
    "permissions": {
        "allow": ["Read(**/*.py)", "Bash(pytest:*)"],
        "deny": ["Delete(**)", "Read(.env*)"]
    },
    "env": {
        "PROJECT_TYPE": "python-fastapi"
    }
}

Templates

Available built-in templates:

• generic - Universal starting point
• web - Frontend applications
• api - Backend services
• cli - Command-line tools
• data-science - ML/Data projects

Best Practices Runbook

1. Run claude-setup init in project root
2. Execute claude-setup configure --auto for detection
3. Review .claude/settings.json permissions
4. Test with claude-setup status --validate
5. Commit .claude/ and CLAUDE.md to git
6. Create project-specific commands in .claude/commands/
7. Add specialized agents to .claude/agents/
8. Document project context in CLAUDE.md
9. Use minimal necessary permissions
10. Update configuration as project evolves

License

MIT License - see LICENSE for details.