Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Using AI assistants with S3 Storage Lens tables

You can use AI assistants and conversational AI tools to interact with your S3 Storage Lens data exported to S3 Tables using natural language. By leveraging the Model Context Protocol (MCP) and the MCP Server for Amazon S3 Tables, you can query, analyze, and gain insights from your storage data without writing SQL queries.

Overview

Model Context Protocol (MCP) is a standardized way for AI applications to access and utilize contextual information. The MCP Server for Amazon S3 Tables provides tools that enable AI assistants to interact with your S3 Tables data using natural language interfaces. This democratizes data access and enables individuals across technical skill levels to work with S3 Storage Lens metrics.

With the MCP Server for S3 Tables, you can use natural language to:

Supported AI assistants

The MCP Server for S3 Tables works with various AI assistants that support the Model Context Protocol, including:

Setting up Kiro with S3 Storage Lens tables

Kiro is an AI coding assistant that provides seamless integration with S3 Tables through the MCP Server. Kiro can help you install and configure the MCP Server directly through its interface, simplifying the setup process.

For more information about Kiro, see Kiro AI.

Prerequisites

Before you begin, ensure that you have:

Step 1: Install the S3 Tables MCP Server

You can install the S3 Tables MCP Server in two ways:

Option 1: Using Kiro's built-in MCP server management

Kiro can help you discover and install MCP servers directly through its interface:

Option 2: Manual installation using uvx

Alternatively, you can manually install the MCP Server using uvx, a Python package runner:

uvx awslabs.s3-tables-mcp-server@latest

For more information about installing the MCP Server, see the Amazon S3 Tables MCP Server documentation.

Step 2: Configure Kiro MCP settings

Create or update your Kiro MCP configuration file at ~/.kiro/settings/mcp.json with the following content:

{
  "mcpServers": {
    "awslabs.s3-tables-mcp-server": {
      "command": "uvx",
      "args": ["awslabs.s3-tables-mcp-server@latest"],
      "env": {
        "AWS_PROFILE": "your-aws-profile",
        "AWS_REGION": "us-east-1"
      }
    }
  }
}

Replace your-aws-profile with your Amazon CLI profile name and us-east-1 with your Amazon Region.

Step 3: Verify the configuration

After configuring the MCP Server, restart Kiro and verify that the S3 Tables tools are available. You can check the available MCP servers in Kiro's settings or by asking Kiro to list available tools.

Example use cases with AI assistants

The following examples demonstrate how to use natural language prompts with AI assistants to interact with S3 Storage Lens data.

Example 1: Query top storage consumers

Prompt: "Show me the top 10 buckets by storage consumption from my S3 Storage Lens data."

The AI assistant will use the MCP Server to query your S3 Storage Lens tables and return the results, including bucket names, storage classes, and storage amounts.

Example 2: Analyze storage growth

Prompt: "Analyze my storage growth over the last 30 days and show me the trend."

The AI assistant will query the storage metrics table, calculate daily storage totals, and present the growth trend.

Example 3: Identify cost optimization opportunities

Prompt: "Find buckets with incomplete multipart uploads older than 7 days that are wasting storage."

The AI assistant will query the storage metrics table for incomplete multipart uploads and provide a list of buckets with potential cost savings.

Example 4: Find cold data candidates

Prompt: "Identify prefixes with no activity in the last 100 days that are stored in hot storage tiers."

The AI assistant will analyze both storage and activity metrics to identify data that could be moved to colder storage tiers for cost optimization.

Example 5: Generate storage reports

Prompt: "Create a summary report of my S3 storage showing total storage, object counts, and request patterns for the last week."

The AI assistant will query multiple tables, aggregate the data, and generate a comprehensive report.

Best practices for using AI assistants

Follow these best practices when using AI assistants with S3 Storage Lens data:

Logging and traceability

When using the S3 Tables MCP Server with AI assistants, you have multiple ways to audit operations:

Security considerations

When using AI assistants with S3 Storage Lens data, follow these security best practices:

Troubleshooting

AI assistant cannot connect to S3 Tables

Problem: The AI assistant reports that it cannot connect to S3 Tables or the MCP Server is not responding.

Solution:

Access denied errors

Problem: The AI assistant receives access denied errors when querying S3 Storage Lens tables.

Solution:

Incorrect or unexpected results

Problem: The AI assistant returns incorrect or unexpected results.

Solution:

Additional resources

For more information about using AI assistants with S3 Tables, see the following resources: