AWS Cost Explorer MCP Server

What does “AWS Cost Explorer” MCP Server do?

The AWS Cost Explorer MCP Server acts as a middleware tool that connects AI assistants, like Anthropic’s Claude, with AWS Cost Explorer and Amazon Bedrock Model Invocation Logs. This server enables developers and AI agents to query and analyze cloud spending data from AWS in natural language, facilitating tasks such as EC2 spend analysis, service spend reports, and granular cost breakdowns. By exposing AWS Cost Explorer API functionality via the Model Context Protocol (MCP), it provides an interactive interface for querying and visualizing AWS costs, which can greatly enhance cloud cost management and reporting workflows. This server can be run locally or remotely, and can aggregate spend data from multiple AWS accounts, provided the correct IAM roles are in place.

List of Prompts

• No explicit prompt templates are listed in the repository or documentation.

List of Resources

• No explicit MCP resources are listed in the repository or documentation.

List of Tools

• No explicit tools or tool names are listed in server.py or README.md regarding MCP tool registration or exposure.

Use Cases of this MCP Server

• EC2 Spend Analysis: Enables developers to get detailed breakdowns of EC2 spending for the previous day, helping identify cost drivers and optimize infrastructure.
• Amazon Bedrock Spend Analysis: Provides insights into Bedrock usage and costs, broken down by region, user, and model, useful for tracking AI/ML workload expenses.
• Service Spend Reports: Allows querying of overall AWS service spending over the last 30 days, aiding in comprehensive cloud cost monitoring.
• Detailed Cost Breakdown: Supports granular analysis of AWS costs by day, region, service, and instance type, empowering precise budget tracking and anomaly detection.
• Cross-account Spend Aggregation: If the IAM role allows, the server can aggregate and report spend from multiple AWS accounts, streamlining multi-account cost management.

How to set it up

Windsurf

1. Ensure Python 3.12, AWS credentials, and Anthropic API access are set up.
2. Locate your Windsurf configuration file.
3. Add the AWS Cost Explorer MCP Server under the mcpServers object:

   {
     "mcpServers": {
       "aws-cost-explorer": {
         "command": "python3",
         "args": ["app.py"]
       }
     }
   }
   

4. Save the config and restart Windsurf.
5. Verify that the server is running and accessible.

Securing API Keys Example:

{
  "mcpServers": {
    "aws-cost-explorer": {
      "command": "python3",
      "args": ["app.py"],
      "env": {
        "AWS_ACCESS_KEY_ID": "your-access-key",
        "AWS_SECRET_ACCESS_KEY": "your-secret-key"
      }
    }
  }
}

Claude

1. Install Python 3.12 and set up AWS credentials.
2. Edit Claude’s MCP configuration file.
3. Add the server as follows:

   {
     "mcpServers": {
       "aws-cost-explorer": {
         "command": "python3",
         "args": ["app.py"]
       }
     }
   }
   

4. Restart Claude Desktop.
5. Confirm integration via the Claude interface.

Cursor

1. Set up Python 3.12 and your AWS credentials.
2. Open Cursor’s configuration file.
3. Insert the following in the mcpServers section:

   {
     "mcpServers": {
       "aws-cost-explorer": {
         "command": "python3",
         "args": ["app.py"]
       }
     }
   }
   

4. Save changes and restart Cursor.
5. Test the connection by running a sample query.

Cline

1. Prepare Python 3.12 and required AWS credentials.
2. Edit the Cline configuration file.
3. Add the server configuration:

   {
     "mcpServers": {
       "aws-cost-explorer": {
         "command": "python3",
         "args": ["app.py"]
       }
     }
   }
   

4. Restart Cline.
5. Ensure the server is operational and responding.

Note: Use environment variables to secure API keys, as shown in the Windsurf example above.

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "aws-cost-explorer": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “aws-cost-explorer” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Our opinion

This MCP server provides a useful interface for AWS cost analytics through Claude and related tools, but lacks explicit MCP prompt, resource, and tool definitions in its documentation. Its setup is straightforward, and it covers a practical cost analysis use case, but some advanced MCP features appear unsupported or undocumented.

MCP Score