Build an AI-powered AWS support companion with Amazon Bedrock AgentCore
Machine Learning Blog
This article demonstrates how to build an AI-powered AWS Support Companion using Amazon Bedrock AgentCore that consolidates incident investigation across multiple AWS tools into a single conversational interface.
- Agent uses Strands Agents framework and Model Context Protocol (MCP) to connect to AWS documentation, Support APIs, and community knowledge from AWS re:Post
- Reduces incident investigation time from 30-45 minutes of context-switching to a unified conversational workflow
- Deploys with single CloudFormation script including Amplify frontend, Cognito authentication, and API Gateway with WAF protection
- Implements security layers: Bedrock Guardrails for content filtering and PII redaction, least-privilege IAM roles, and CloudTrail audit logging
- Agent code demonstrates parallel MCP initialization, memory with graceful timeouts, and automatic token refresh for production reliability
- Supports queries for AWS best practices, troubleshooting guidance, and automated support case creation
The solution enables AWS teams to investigate incidents faster by eliminating manual console switching and consolidating investigation tools into a single AI-powered interface.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
2026
2026
2026
2026
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.