AWS has introduced an AI-powered support companion designed to streamline the process of investigating and resolving infrastructure issues. The tool, built using Amazon Bedrock AgentCore, consolidates multiple manual steps into a single conversational interface, allowing engineers to analyze CloudWatch logs, search documentation, and create support cases without switching tools. This integration aims to reduce the time spent on each incident investigation by 30–45 minutes, according to AWS.
The solution uses Amazon Bedrock AgentCore to manage the operational complexity of running a production AI agent, including session isolation, auto scaling, security, and observability. It connects to AWS services through the Model Context Protocol (MCP), which provides a standard way for AI agents to access external tools. The agent can analyze logs, query community knowledge from AWS re:Post, and create support cases, all through a single interface. The deployment process uses AWS CloudFormation and includes a web frontend built on AWS Amplify for user interaction.
The solution addresses the repetitive and time-consuming nature of manual incident investigation, which involves checking CloudWatch logs, searching documentation, and reviewing community posts. AWS emphasized that the tool eliminates the need to switch between multiple tools and interfaces, thereby improving response speed and reducing context loss during troubleshooting. The deployment process includes setting up infrastructure, building a Docker container, and configuring the frontend for user access.
Source: awsml
