sre-mcp-server

Clovhis/sre-mcp-server

3.2

If you are the rightful owner of sre-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Model Context Protocol (MCP) server is a specialized server designed to facilitate communication and data exchange between various components in a machine learning or AI-driven environment. It leverages the MCP framework to ensure seamless integration and efficient data handling.

sre-mcp-server

Pre-requisites

Before starting, ensure you have a compatible version of the Python interpreter installed. It is recommended to use pyenv to manage and install the required Python version. Additionally, the project relies on the uv package and project management tool, which should be installed prior to setup.

Getting Started

After cloning the repository, navigate to the project directory and run:

uv sync

This command will create a virtual environment and install all dependencies specified in pyproject.toml.

Configuration

Rename the .env.dev file to .env:

mv .env.dev .env

Edit the .env file to set the Azure authentication environment variables required for the MCP tools:

AZURE_TENANT_ID=<your-azure-tenant-id>
AZURE_CLIENT_ID=<your-azure-client-id>
AZURE_CLIENT_SECRET=<your-azure-client-secret>
# Azure Search
SEARCH_ENDPOINT="https://<your-search-endpoint>.search.windows.net"
SEARCH_INDEX_NAME="<your-search-index-name>"
SEARCH_API_KEY="<your-search-api-key>"
# API Keys
API_KEYS="<your-mcp-api-keys>"

Replace the placeholder values with your actual Azure credentials.

Running MCP Servers

An example MCP server has been configured in .vscode/mcp.json along with the Azure MCP server. To run them, open the Command Palette and execute the MCP: List Servers command. Select the desired server from the list to start it.

Alternatively, you can start the servers directly from .vscode/mcp.json. Each MCP server entry includes a Start button at the top, allowing you to launch the server with a single click.

Testing in Copilot Agent Mode

Before testing the MCP server in Copilot Agent mode, ensure you are authenticated with Azure. Run the following command and complete the login process in your browser:

az login --use-device-code

This step is required for the MCP server to access Azure resources during testing.

Using Copilot Agent Mode

To use Copilot Agent mode with this project:

  1. Ensure the MCP server is running as described above.
  2. In your Copilot-enabled IDE, set the agent endpoint to point to your running MCP server (refer to your IDE's documentation for configuring agent endpoints).
  3. Interact with the Copilot Agent as you normally would; requests will be routed through your MCP server.

For troubleshooting or advanced configuration, consult the project's documentation or open an issue for support.

Smart Automation

**Subscription name:** <subscription-name>
**Subscription ID:** <subscription-id>

To analyze average CPU usage growth in Kubernetes container instances for log analytics workspaces (specifically those starting with `USN`)