clauderc
Health Warn
- No license — Repository has no license file
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Community trust — 15 GitHub stars
Code Fail
- process.env — Environment variable access in bin/cli.js
- execSync — Synchronous shell command execution in src/detector.js
- execSync — Synchronous shell command execution in src/project.js
Permissions Pass
- Permissions — No dangerous permissions requested
This tool is a CLI utility designed to quickly configure local and project-level environments for Claude Code. It installs pre-made best-practice templates, custom slash commands, and agent profiles.
Security Assessment
The overall security risk is Medium. The tool actively uses synchronous shell command execution (`execSync`) in its detector and project setup scripts. While normal for a CLI scaffolding tool that modifies local directories and runs commands, executing shell instructions always carries a risk. If the package is compromised or scripts are altered, it could lead to arbitrary command execution on your machine. The code accesses environment variables, but no hardcoded secrets or dangerous permissions were found.
Quality Assessment
The project appears actively maintained, with repository activity as recent as today. It has a small but positive community trust indicator with 15 GitHub stars. However, the automated scan flags a missing license file in the repository (even though the README claims it uses the MIT license). Always review the repository directly to confirm licensing terms before adopting it in a commercial environment.
Verdict
Use with caution. While it seems to be a helpful and active setup tool, verify the shell execution scripts in the source code to ensure you are comfortable with the commands it runs on your system.
Setup Claude Code with best practices - agents, skills, commands, and templates.
clauderc
Setup Claude Code with best practices - agents, skills, commands, and templates.
Quick Start • What's Included • Commands • Best Practices
Based on my experiences and tips from Boris Cherny (creator of the Claude Code) and official documentation.
Quick Start
# Global setup (agents, skills, commands, templates)
npx clauderc init
# Project setup (CLAUDE.md, .claude/, settings)
npx clauderc project
# Update to latest version
npx clauderc update
Works on macOS, Linux, and Windows.
Two-Level Setup
Global (~/.claude/)
Shared components available to all projects:
npx clauderc init
~/.claude/
├── agents/
│ └── project-setup-wizard.md # Setup Claude Code for any project
├── skills/
│ ├── project-analysis/ # Systematic project analysis
│ ├── claude-code-templates/ # Quick reference for templates
│ ├── evolve-claude-md/ # Self-improving CLAUDE.md
│ └── explain/ # Codebase visualization
├── commands/
│ ├── test.md # /test - run tests
│ ├── lint.md # /lint - lint and format
│ ├── verify.md # /verify - full validation
│ ├── pr.md # /pr - create pull request
│ ├── setup.md # /setup - install dependencies
│ ├── commit.md # /commit - semantic commits
│ ├── worktree.md # /worktree - parallel workspaces
│ ├── fix.md # /fix - autonomous bug fixing
│ └── review.md # /review - code review
└── templates/project-setup/
└── ... # Reference templates
Project (.claude/ + CLAUDE.md)
Project-specific configuration:
cd your-project
npx clauderc project
your-project/
├── CLAUDE.md # Project context for Claude
└── .claude/
├── settings.json # Permissions & hooks
└── commands/ # Project-specific commands
├── dev.md
├── test.md
├── lint.md
└── ...
Supported Stacks
Auto-detection for 11+ languages:
| Stack | Detection | Package Managers |
|---|---|---|
| Node.js/TypeScript | package.json |
npm, pnpm, yarn, bun |
| Python | pyproject.toml, requirements.txt |
poetry, pipenv, uv, pip |
| Go | go.mod |
go mod |
| Rust | Cargo.toml |
cargo |
| Java/Kotlin | pom.xml, build.gradle |
maven, gradle |
| PHP | composer.json |
composer |
| Ruby | Gemfile |
bundler |
| C#/.NET | *.csproj, *.sln |
dotnet |
| Elixir | mix.exs |
mix |
| Swift | Package.swift |
swift |
| Dart/Flutter | pubspec.yaml |
pub, flutter |
Also detects: Monorepos (Turborepo, Nx, Lerna), CI/CD (GitHub Actions, GitLab CI, etc.)
Commands
init
Install global components to ~/.claude/:
npx clauderc init
# Force overwrite existing files
npx clauderc init --force
# Preview changes without applying
npx clauderc init --dry-run
project
Setup current project with interactive wizard:
npx clauderc project
Features:
- Auto-detects stack, package manager, framework
- Generates appropriate
CLAUDE.md - Creates
.claude/settings.jsonwith permissions - Creates project-specific commands
update
Update global components to latest version:
npx clauderc update
# Preview what would be updated
npx clauderc update --dry-run
Features:
- Automatic backup of modified files
- Shows changelog between versions
- Only updates files that changed
list
Show installed components:
npx clauderc list
changelog
Show version history:
npx clauderc changelog
Usage in Claude Code
After installation:
# Setup wizard for any project
"Use the project-setup-wizard agent to configure Claude Code for this project"
# Quick commands
/test # Run project tests
/lint # Lint and format code
/verify # Full verification (lint + test + build)
/pr # Create a pull request
/setup # Install dependencies
/commit # Semantic commits
/worktree # Manage parallel workspaces
/fix # Autonomous bug fixing
/review # Code review
# Skills
"Use project-analysis skill to analyze this codebase"
Best Practices (from Boris Cherny)
- Multiple instances - Run 5+ Claudes in parallel for independent tasks
- Opus 4.6 - Best model for coding, use
/fastfor faster output - Shared CLAUDE.md - Team commits and updates regularly
- Plan Mode - Use for complex tasks (Shift+Tab twice)
- Slash commands - Automate repetitive workflows
- PostToolUse hooks - Auto-format code after edits
- Permissions - Pre-authorize safe commands instead of
--dangerously-skip-permissions - Feedback loop - Give Claude a way to verify its work
Contributing
Contributions are welcome! Please read our guidelines before contributing:
- Contributing Guide - How to contribute to the project
- Code of Conduct - Community guidelines
- Security Policy - How to report vulnerabilities
Ways to contribute:
- Add new commands, skills, or agents
- Add support for new stacks/languages
- Improve existing templates
- Report bugs or suggest features
- Improve documentation
Author
Matheus Kindrazki
- GitHub: @matheuskindrazki
- Twitter: @kindraScript
License
MIT
⭐ If this helped you, consider giving it a star!
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found