MCP Agent Orchestration System

MCP Agent Orchestration System

A Python implementation of a state-based agent orchestration system using the Model Context Protocol (MCP).

What is MCP?

The Model Context Protocol (MCP) allows applications to provide context for LLMs in a standardized way, separating the concerns of providing context from the actual LLM interaction. With MCP, you can build servers that expose:

• Resources: Data sources that provide information to LLMs
• Tools: Functions that allow LLMs to perform actions
• Prompts: Reusable templates for LLM interactions

Installation

Prerequisites

• Python 3.10 or higher
• MCP Python SDK 1.2.0 or higher

Setting Up Your Environment

Using uv (recommended)

Using pip

Clone or Download Project Files

Place the project files in your directory:

• orchestrator.py - The main MCP server implementing the state machine
• orchestrator_client.py - Client demonstrating the orchestration flow
• requirements.txt - Dependencies for the project
• .gitignore - Git ignore file

Project Structure

• orchestrator.py - The main MCP server implementing the state machine
• orchestrator_client.py - Client demonstrating the orchestration flow
• requirements.txt - Dependencies for the project

Running the Orchestration System

1. Start the orchestration server directly for testing:

2. In a separate terminal, run the client to see the orchestration in action:

Integrating with Claude for Desktop

1. Install Claude for Desktop

Make sure you have Claude for Desktop installed. You can download the latest version from Anthropic's website.

2. Configure Claude for Desktop

1. Open your Claude for Desktop configuration file:macOS/Linux:
   # Create or edit the configuration file code ~/Library/Application\ Support/Claude/claude_desktop_config.json
   Windows:
   # Path may vary depending on your Windows version code %APPDATA%\Claude\claude_desktop_config.json
2. Add the orchestrator server configuration:
   { "mcpServers": { "agent-orchestrator": { "command": "python", "args": [ "/ABSOLUTE/PATH/TO/YOUR/PROJECT/orchestrator.py" ] } } }
   Replace the path with the absolute path to your orchestrator.py file.
3. Save the configuration file and restart Claude for Desktop.

3. Using the Orchestrator in Claude

Once configured, you can:

1. Open Claude for Desktop
2. Click on the MCP server icon in the sidebar
3. Select "agent-orchestrator" from the list of available servers
4. Start interacting with the orchestration system

Claude will be able to:

• Transition between different agent states
• Store and retrieve information from the knowledge base
• Maintain conversation context across state transitions
• Access state-specific prompts

Agent States

The orchestration system implements a state machine with the following states:

• IDLE: Waiting for instructions
• PLANNING: Creating a structured plan for a task
• RESEARCHING: Gathering information needed for a task
• EXECUTING: Carrying out planned actions
• REVIEWING: Evaluating results and determining next steps
• ERROR: Handling errors or unexpected situations

Customizing the System

Adding New States

1. Add the state to the AgentState enum in orchestrator.py
2. Create a prompt function for the new state
3. Update the transition logic in _get_available_transitions()
4. Add handlers for the new state in resource access functions

Creating Custom Tools

Add new tools by creating functions decorated with @mcp.tool():

Development and Testing

Using the MCP CLI

The MCP CLI provides tools for development and testing:

Manual Testing with Python

Resources

• MCP Python SDK Documentation
• Model Context Protocol Specification

License

This project is licensed under the MIT License - see the LICENSE file for details.