mcpserver.so LogoMCPFly
Submit
MCP Servers Collection logo

MCP Servers Collection

by mows21

AI/Automation & WorkflowAI AgentsAutomationClineRooClaude DesktopDockerJSON-RPC

A collection of Model Context Protocol (MCP) servers designed for Cline, Roo, and Claude Desktop. These servers enable AI assistants to interact with local and remote resources through a standardized protocol.

View on GitHub

Last updated: N/A

MCP Servers Collection

A collection of Model Context Protocol (MCP) servers designed for Cline, Roo, and Claude Desktop.

Overview

This repository contains a comprehensive suite of MCP servers that enable AI assistants to interact with local and remote resources through a standardized protocol.

Available Servers

  1. Cline MCP Server (Port 8080)

    • Internal workflow orchestration
    • Agent task execution
    • Memory management
    • System execution capabilities

  2. Roo MCP Server (Port 8081)

    • Web scraping and automation
    • Browser control with Playwright
    • Form filling automation
    • Screenshot/PDF capture
    • Watchdog sessions

  3. Claude Desktop MCP Server (Port 8082)

    • Local file operations
    • Shell command execution
    • Desktop notifications
    • LLM integration (Ollama)
    • System information access

Quick Start

Using Docker

# Clone the repository
git clone https://github.com/mows21/mcp-servers.git
cd mcp-servers

# Start all servers
docker-compose up -d

# Check server health
docker-compose ps

# View logs
docker-compose logs -f

Manual Installation

# Install requirements for each server
cd cline-mcp && pip install -r requirements.txt
cd ../roo-mcp && pip install -r requirements.txt
cd ../claude-desktop-mcp && pip install -r requirements.txt

# Start servers individually
cd cline-mcp && python main.py
cd ../roo-mcp && python main.py
cd ../claude-desktop-mcp && python main.py

Configuration

Claude Desktop Integration

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "cline-internal": {
      "command": "docker",
      "args": ["exec", "-i", "cline-mcp-server", "python", "main.py"]
    },
    "roo-automation": {
      "command": "docker",
      "args": ["exec", "-i", "roo-mcp-server", "python", "main.py"]
    },
    "claude-desktop-local": {
      "command": "docker",
      "args": ["exec", "-i", "claude-desktop-mcp-server", "python", "main.py"]
    }
  }
}

Architecture

The MCP servers communicate through:

  • Unified /mcp endpoints
  • JSON-RPC protocol
  • Docker network isolation
  • Event-driven architecture

API Documentation

Cline MCP Server

  • POST /agent/run - Execute agent tasks
  • POST /workflow - Run workflows
  • GET/POST/DELETE /memory - Memory management
  • POST /exec - System command execution

Roo MCP Server

  • POST /scrape/url - Web scraping
  • POST /form/fill - Form automation
  • POST /watchdog - Session monitoring
  • POST /snapshot - Page capture

Claude Desktop MCP Server

  • POST /ask - LLM interaction
  • GET /file/read - File reading
  • POST /file/write - File writing
  • POST /shell/run - Shell execution
  • POST /notify - Desktop notifications

Security

  • Network isolation through Docker
  • Token-based authentication
  • Input validation and sanitization
  • Resource limits and health checks

Contributing

  1. Fork the repository
  2. Create your feature branch
  3. Commit your changes
  4. Push to the branch
  5. Create a Pull Request

License

MIT License - See LICENSE file for details

Contact

For questions and support, please open an issue on GitHub.

Acknowledgments

Built for the Model Context Protocol ecosystem to enhance AI agent capabilities.