claude-code-mcp-init
A lightweight CLI tool to easily configure and initialize MCPs for Claude Code.
GitHubスター
60
ユーザー評価
未評価
お気に入り
0
閲覧数
13
フォーク
4
イシュー
1
Claude Code MCP Configuration Manager
⭐ This repo helped you? Consider giving it a star! ⭐
A lightweight CLI for configuring Model Context Protocol (MCP) servers with Claude Code.
Table of Contents
- Features
- Supported MCP Servers
- Quick Setup
- Commands
- Environment Variables
- File Structure
- Troubleshooting
- Contributing
Features
- 🎨 Beautiful CLI Interface - Colorful, intuitive menus and status displays
- 🔧 Interactive Configuration - Choose which MCP servers to configure
- ✅ Requirement Validation - Automatically checks if API keys and dependencies are available
- 🚀 One-Command Setup - Install and configure everything with simple commands
- 🔄 Server Management - List, remove, and check status of configured servers
- 🔐 Secure Configuration - Environment-based API key management
Current Supported MCP Servers
Server | Description | Requirements |
---|---|---|
AWS CloudWatch | Access AWS logs and metrics | AWS_PROFILE , AWS_REGION |
Snowflake | Query databases with natural language | SNOWFLAKE_ACCOUNT , SNOWFLAKE_USER , etc. |
Linear | Project management integration | None (remote server) |
GitHub | Repository and issue management | GITHUB_PAT or GITHUB_TOKEN |
Notion | Documentation and knowledge base | NOTION_API_KEY |
Tavily | Web search and content extraction | TAVILY_API_KEY |
Zen | Enhanced AI capabilities and workflows | OPENAI_API_KEY |
MongoDB | Database operations | MDB_MCP_CONNECTION_STRING |
Context7 | Up-to-date code documentation for any prompt | None (remote server or npm package) |
Interrupt User | Ask human for context and additional information | None (uvx package) |
Quick Setup
Clone and install
git clone https://github.com/Eyalbenba/claude-code-mcp-init cd claude-code-mcp-init ./src/mcp-setup/install-claude-mcp.sh
Configure environment
cp .env-mcp.example .env-mcp # Edit .env-mcp with your API keys
Setup MCP servers
claude-mcp status # Check requirements claude-mcp configure # Interactive setup
Commands
Command | Description |
---|---|
claude-mcp configure |
Interactive MCP server configuration |
claude-mcp status |
Check server requirements and status |
claude-mcp list |
List configured MCP servers |
claude-mcp remove |
Remove MCP servers |
claude-mcp help |
Show help information |
Troubleshooting
Command not found
# Restart terminal or add to PATH
export PATH="$HOME/.local/bin:$PATH"
Bash version issues (macOS)
brew install bash
/opt/homebrew/bin/bash claude-mcp configure
Missing API keys
The claude-mcp status
command shows exactly which environment variables are missing.
Contributing
We welcome contributions! 🎉 Here's how to get started:
Quick Start: Fork → Clone → Test → Submit PR
🚀 Ways to Contribute
- 🐛 Bug fixes - Fix issues or improve error handling
- ✨ New MCP servers - Add support for more services
- 📚 Documentation - Improve guides and examples
- 🧪 Testing - Help validate configurations
📋 Contribution Standards
- Branch naming:
feat/server-name
orbugfix/description
- Test your changes: Run
claude-mcp status
andclaude-mcp configure
- Follow patterns: Use existing code style in
src/mcp-setup/claude-mcp
- Update docs: Add new servers to README table and
.env-mcp.example
🔧 Adding a New MCP Server
Step-by-Step Guide
Define the server in
src/mcp-setup/claude-mcp
:MCP_SERVERS[newserver]="New Server Name" MCP_DESCRIPTIONS[newserver]="Description of what this server does" MCP_REQUIREMENTS[newserver]="REQUIRED_ENV_VAR, ANOTHER_VAR"
Add configuration function:
configure_newserver_server() { if claude mcp add "server-name" "$UVX_PATH" \ server-package@latest \ -e "ENV_VAR=$ENV_VAR" \ -t stdio; then print_success "New Server configured successfully" else print_error "Failed to configure New Server" return 1 fi }
Wire it up in the case statement:
case "$server_key" in # ... existing servers ... "newserver") configure_newserver_server ;;
Update documentation:
- Add server to the Supported MCP Servers table above
- Add environment variables to
.env-mcp.example
Test thoroughly:
claude-mcp status # Should show your new server claude-mcp configure # Should allow configuring it
💻 Development Setup
Quick Setup
# Fork the repo on GitHub, then:
git clone https://github.com/YOUR-USERNAME/claude-mcp-init
cd claude-mcp-init
./src/mcp-setup/install-claude-mcp.sh
# Create test environment
cp .env-mcp.example .env-mcp
# Edit .env-mcp with your test credentials
# Test your changes
claude-mcp status
claude-mcp configure
Requirements
- Bash 4.0+ (auto-installed on macOS via Homebrew)
- uvx (auto-detected or set
UVX_PATH_OVERRIDE
) - Claude Code CLI
📝 Pull Request Checklist
Before Submitting
- Tested locally - Your changes work with
claude-mcp status
andclaude-mcp configure
- Documentation updated - README table and
.env-mcp.example
if adding new server - Code follows patterns - Matches existing style and structure
- Branch named correctly -
feat/server-name
orbugfix/description
- Clear commit message - Explains what and why
Pull Request Template
## Changes
- Brief description of what you added/fixed
## Testing
- [ ] Tested with `claude-mcp status`
- [ ] Tested with `claude-mcp configure`
- [ ] Updated documentation if needed
## Type of Change
- [ ] Bug fix
- [ ] New MCP server
- [ ] Documentation update
- [ ] Other: ___________
🤝 Community Guidelines
Code of Conduct
- Be respectful and inclusive in all interactions
- Help others learn and contribute
- Focus on constructive feedback in reviews
- Assume good intentions from contributors
Getting Help
- Issues - Use GitHub Issues for bugs and feature requests
- Discussions - Use GitHub Discussions for questions
- Security - Email security issues privately (see Security section)
Recognition
All contributors will be recognized in our release notes and contributor acknowledgments.
A multi-agent AI architecture that connects 25+ specialized agents through n8n and MCP servers. Project NOVA routes requests to domain-specific experts, enabling control of applications from knowledge bases to DAWs, home automation to development tools. Includes system prompts, Dockerfiles, and workflows for a complete AI assistant ecosystem.
A multi-agent AI architecture that connects 25+ specialized agents through n8n and MCP servers. Project NOVA routes requests to domain-specific experts, enabling control of applications from knowledge bases to DAWs, home automation to development tools. Includes system prompts, Dockerfiles, and workflows for a complete AI assistant ecosystem.