This MCP Server Collection provides multiple MCP-based services that you can run and interact with using a consistent protocol. The HTTP Streaming MCP server lets you spawn a streamable HTTP interface for agents to interact with external services in real time. The Google Calendar MCP server exposes operations to list, create, delete, and update calendar events, enabling agents to manage calendars through MCP. The File System Explorer MCP provides a simple API to perform file and directory operations, useful for agents that need to read, write, or manipulate local or mounted files. The Azure OpenAI Client MCP connects to Azure OpenAI endpoints, allowing agents to send prompts and receive completions via Azure’s AI services. To use a particular server, start it with its Python script and then use your MCP client tooling to connect to the server’s endpoint following the MCP protocol described by the repository’s examples. Each server exposes its own capabilities via MCP messages; for instance, you can request calendar events, file listings, or OpenAI completions, and receive structured MCP responses suitable for agent orchestration.

Prerequisites:

• Python 3.8+ installed on your system
• Git to clone the repository
• Optional: virtual environment tools (venv, virtualenv)

Steps:

1. Clone the repository: git clone https://github.com/Pranesh-2005/MCP_Server_Collections.git cd MCP_Server_Collections

2. Create and activate a virtual environment (optional but recommended): python -m venv venv

   Windows

   venv\Scripts\activate

   macOS/Linux

   source venv/bin/activate

3. Install per-submodule dependencies:

   • For HTTP Streamable MCP Server (if a requirements file exists, otherwise skip): cd HTTP-MCP pip install -r requirements.txt # if present cd ..

   • For Google Calendar MCP Server: cd MCP-Calendar

     Check the README for exact requirements; example from repo includes:

     pip install fastmcp google-auth google-auth-oauthlib google-api-python-client python-dotenv openai cd ..

   • For File System Explorer MCP: cd MCP-File_System

     Check the README for exact requirements; install as appropriate

     cd ..

   • For Azure OpenAI Client MCP: cd MCP-Client-Azure

     Install any required dependencies listed in its README

     cd ..

4. Run the servers you want to test:

   • HTTP Streaming MCP Server: python HTTP-MCP/server.py

   • Google Calendar MCP Server: python MCP-Calendar/calendar_mcp_server.py

   • File System Explorer MCP: python MCP-File_System/filesystem_mcp_server.py

   • Azure OpenAI Client MCP:

     Ensure AZURE_OPENAI_ENDPOINT and AZURE_OPENAI_API_KEY are set in your environment or .env

     python MCP-Client-Azure/client.py

Note: Each submodule may have its own specific prerequisites and authentication steps (e.g., Google API credentials, Azure credentials). Refer to each submodule’s Readme.md for precise installation and setup instructions.

Tips and tips:

• If running multiple servers locally, ensure they listen on distinct ports or use a gateway to route MCP messages to the correct service.
• For Google Calendar integration, you’ll typically need OAuth credentials; prepare a credentials.json or set the necessary environment vars as described in the specific submodule docs.
• The Azure OpenAI client requires endpoint and API key. Do not expose these in code; use environment variables or a secret manager.
• If you encounter module-not-found errors, re-check the per-submodule requirements and ensure you’re in the correct virtual environment where dependencies were installed.
• The repository mentions GitHub automation tools; if you intend to enable those, follow the MCP-GITHUB documentation to set up any required tokens and permissions.