Multi Tools

[!NOTE] Model Context Protocol (MCP) Bridge - A powerful service that connects AI agents with DevOps and communication tools through a standardized interface.

Welcome to the MCP Server Multi Tools! This service acts as a seamless integration layer, enabling AI agents and systems to interact with services like Azure DevOps, Slack, GitHub, and even other agents (with their own tools) using a simple, unified protocol.

---

🛠️ Available Tools & Providers

📊 DevOps & Project Management

A comprehensive suite for total project management:

• 📝 Work Items: Create, read, update, search, and comment
• 🏃‍♂️ Sprints: Manage sprints, view contents, and track progress
• 🔍 WIQL: Execute custom Work Item Query Language statements
• 🔗 Enrichment: Augment work items with GitHub, Slack, and Sentry context
• 📁 Git Content: Fetch file content from associated repositories

🐙 GitHub Integration

• Seamless version control and pull request workflows
• Repository content access and management

💬 Communication

• Real-time notifications and messaging
• Channel posting for agent communication
• Team collaboration enhancement

🤖 Advanced Agent System

[!IMPORTANT] Agents-in-Agents: Revolutionary nested agent architecture

Your AI agent can create and manage other agents, each with:

• 🐋 Docker container with full Debian Linux system
• 🌐 Web browser capabilities
• 🔄 Iterative work processes
• 💬 Inter-agent communication

---

🚀 Quick Start Guide

📋 Prerequisites

[!WARNING] Ensure you have the following before proceeding:

• ✅ Go (latest version)
• 🔑 Service credentials (Azure DevOps, Slack tokens, etc.)
• 🐳 Docker Desktop (for agents-in-agents feature)

Step 1️⃣: Clone & Build

# Clone the repository
git clone https://github.com/theapemachine/mcp-server-multi-tools.git
cd mcp-server-devops-bridge

# Build the server
go build -o mcp-server-multi-tools .

Step 2️⃣: Environment Setup

# Copy and configure environment variables
cp start.sh.example start.sh
vim start.sh  # Add your credentials

[!TIP] The server is configured entirely through environment variables for maximum flexibility.

Step 3️⃣: MCP Client Configuration

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "multi-tools": {
      "command": "/path/to/mcp-server-multi-tools/start.sh",
      "args": []
    }
  }
}

---

🎬 See It In Action

[!NOTE] Real AI Development Workflow - Watch Claude Sonnet 4 leverage the bridge to understand requirements, manage tasks, and write code.

The agent uses MCP tools to orchestrate complex development workflows through a single, unified interface.

For a complete history of testing, including the upgraded communication system and the headless browser tool, see agents-full.pdf.

---

🤝 Contributing

We welcome contributions! Here's how to get involved:

🔄 Contribution Workflow

1. 🍴 Fork the repository
2. 🌿 Branch your feature (git checkout -b feature/AmazingFeature)
3. 💾 Commit your changes (git commit -m 'Add AmazingFeature')
4. 🚀 Push to branch (git push origin feature/AmazingFeature)
5. 📬 Open a Pull Request

💡 What We're Looking For

• 🔧 New tool integrations
• 📚 Documentation improvements
• 🐛 Bug fixes and optimizations
• 🎨 UI/UX enhancements

---

📜 License

MIT License - See LICENSE for details

---

Built with 🤷 for the developer community