Getting started with the AWS Support MCP and Amazon Q CLI

In today's development landscape, developers increasingly rely on Amazon Q for debugging issues, leveraging its powerful ability to parse through documentation and provide relevant solutions. However, when challenges require deeper investigation or access to underlying AWS resources, AWS Support becomes essential. By combining Amazon Q with the Support MCP server, developers can streamline their support workflow significantly. This integration allows them to seamlessly create and manage support cases while maintaining the context of their conversation with the AI assistant, eliminating the need to repeatedly explain their technical issues. The result is a more efficient, integrated support experience that bridges the gap between AI-assisted troubleshooting and expert AWS Support. Here's how to get started: Prerequisites:

• Python 3.7 or higher
• AWS credentials with Support API access
• Active Business, Enterprise On-Ramp, or Enterprise Support plan
• uv package manager (install from Astral or GitHub)

Step 1: Set Up Your Environment

First, install Python using uv:

uv python install 3.10

Step 2: Install the AWS Support MCP Server

Create or edit your global MCP client configuration file at ~/.aws/amazonq/mcp.json:

`{
    "mcpServers": {
      "awslabs_support_mcp_server": {
         "command": "uvx",
         "args": [
            "-m", "awslabs.aws-support-mcp-server@latest",
            "--debug",
            "--log-file",
            "./logs/mcp_support_server.log"
         ],
         "env": {
            "AWS_PROFILE": "your-aws-profile"
         }
      }
   }
}`

Step 3: Optional - Configure Support MCP at a Workspace or local level only

You can add MCPs to individual workspaces so your assistant can only access certain MCPs from within projects with the following command:

qchat mcp add --name awslabs_support_mcp_server --command uvx

Amazon Q will by default create an ./amazonq/ directory to be used for storing local workspace context. Be aware if both local and global configuration files exist, Amazon Q reads MCP configuration from both and combines them, taking the union of their contents. If there is a conflict (i.e., a server defined in the global config is also present in the workspace config), a warning is displayed and only the server entry in the workspace config is used.

What Can You Do? Once configured, your AI assistant can:

• Create and manage AWS support cases
• Retrieve case information and communications
• Add communications to existing cases
• Resolve support cases
• Determine appropriate issue types, service codes, and category codes
• Set appropriate severity levels for cases

Now your AI assistant is ready to interact with AWS Support services through the MCP server!

References:

https://github.com/awslabs/mcp/tree/main/src/aws-support-mcp-server

https://docs.astral.sh/uv/getting-started/installation/

https://docs.aws.amazon.com/amazonq/latest/qdeveloper-ug/command-line-mcp-understanding-config.html