MCPO Project

This repository contains the source code, configuration files, and deployment instructions for a project that uses Docker, Kubernetes, and the mcpo tool to manage mulitiple MCP (Multi-Component Protocol) services.

Overview

The project sets up:

• A Docker container running the mcpo tool with configuration from mcpo_config.yaml managed by config_builder.py
• A Kubernetes deployment for the Gitea MCP application
• Services and pods to manage the application's networking and lifecycle

Key Components

Docker

• Dockerfile: Builds a container with Python dependencies and the mcpo tool
• gitea_mcp_set.sh: Script to configure gitea_mcp in the container environment
• requirements.txt: Python dependencies for any script-based components

Kubernetes

• kube/deployment.yaml: Defines the Kubernetes deployment for the Gitea MCP application
• kube/pod.yaml: Pod specification for the container
• kube/service.yaml: Service configuration to expose the application

Configuration

Configuration

• Configuration File: Managed through mcpo_config.yaml.
• Conversion Tool: Uses config_builder.py to convert YAML to JSON.
• Template Guidelines: Assumes the yaml will follw the guideline template outlined in the MCPo configuration guidelines. It doesn't perform validation so it won't fail until build if you fuck it up

Purpose

This project provides a complete environment to:

1. Run the mpco service using containerization
2. Orchestrate the service using Kubernetes
3. Manage configuration and authentication securely

Setup Instructions

1. Clone the repository:

   git clone [repository-url]
   

2. Build the Docker image:

   docker build -t mcpo-service:latest .
   

3. Apply Kubernetes configurations:

   kubectl apply -f kube/
   

4. Verify the deployment:

   kubectl get pods -n mcpo
   kubectl get services -n mcpo
   

5. Test the service:

   • Ensure the service is accessible through the NodePort (30026)
   • Verify the MCP servers are configured correctly

Notes

• The mcpo tool is used to manage MCP connections
• Sensitive information should be parsed via environment variables to mcpo_config.yaml
• mcpo_config.yaml will parse env vars using the following syntax !ENV ${YOUR_VAR}
• The project is designed for production deployment with Kubernetes orchestration

For detailed explanations of individual components, refer to their respective files in the repository.