AWS DevOps Agent FAQs

AWS DevOps Agent is a frontier agent that resolves and proactively prevents incidents, continuously improving reliability and performance. AWS DevOps Agent investigates incidents and identifies operational improvements as an experienced DevOps engineer would: by learning your resources and their relationships; working with your observability tools, runbooks, code repositories, and CI/CD pipelines; and correlating telemetry, code, and deployment data across all of them to understand the relationships between your application resources, including applications in multicloud and hybrid environments. DevOps Agent uses this deep understanding of your operations and workloads to reduce MTTR and drive operational excellence.

Frontier agents represent a new class of AI agents. They're autonomous systems that work independently to achieve goals, scale massively to tackle concurrent tasks, and run persistently for hours or days without intervention. Unlike traditional AI assistants that help with individual tasks, frontier agents act as extensions of your team, delivering complete outcomes across diverse use cases.

AWS DevOps Agent autonomously triages operational incidents as they arise and guides teams to rapid resolution based on a deep understanding of your operations and workloads. It moves teams from reactive firefighting to proactive operational improvement by analyzing patterns across historical incidents to deliver targeted recommendations that prevent future incidents and strengthen system resilience. AWS DevOps Agent continually refines its recommendations, aligned with your operational priorities, to keep improving the reliability, performance, and efficiency of your applications based on your team's feedback.

Using the investigations capability in Amazon CloudWatch, you can accelerate operational investigations across your AWS environment, correlating data from AWS services including CloudWatch telemetry, AWS CloudTrail Logs, deployment information, changes to resource configuration, and AWS Health events. CloudWatch investigations is now generally available at no additional cost. AWS DevOps Agent extends beyond AWS services. Add AWS DevOps Agent to your team to investigate and identify operational improvements. It can connect to your AWS and third-party observability tools, runbooks, code repositories, and CI/CD pipelines, correlating telemetry, code, and deployment data across all of them to understand the relationships between your application components, including applications in multicloud and hybrid environments. AWS DevOps Agent is currently available in Preview.

Yes, AWS DevOps Agent can connect to AWS and third-party observability tools to introspect relevant alarms, metrics, logs, and traces contained in your existing observability tools. AWS DevOps Agent offers built-in integrations with observability tools such as Amazon CloudWatch, Dynatrace, Datadog, New Relic, and Splunk. You can also connect to other observability tools by connecting to your own MCP server.

Yes, AWS DevOps Agent can connect to AWS and third-party pipeline tools to introspect deployments and code diffs as it searches for an incident’s root cause. AWS DevOps Agent offers built-in integrations with CI/CD pipelines like GitHub and GitLab. You can also connect to other tools by connecting to your own MCP server.

Yes, AWS DevOps Agent can connect to your ticketing tools to trigger root cause investigations from tickets or alarms and publish its findings and suggested mitigation plans back to the originating ticket. You can also connect AWS DevOps Agent to your team’s collaboration tools to enable it to publish its activity and findings in team collaboration channels.

AWS DevOps Agent offers built-in integrations with observability tools such as Amazon CloudWatch, Dynatrace, Datadog, New Relic, and Splunk, and CI/CD pipelines like GitHub and GitLab. You can extend AWS DevOps Agent beyond its built-in integrations by connecting to your own MCP server, enabling integrations with additional tools such as your organization’s custom tools, specialized platforms, or proprietary ticketing systems. AWS DevOps Agent also integrates with collaboration tools such as Slack, ServiceNow, and PagerDuty, streamlining incident response coordination.

We are not charging for AWS DevOps Agent during the preview period (with limits on the number of agent task hours per month), but you are responsible for fees incurred for other AWS services that you use and connect to AWS DevOps Agent.

Your account will be limited to 10 Agent Spaces and 20 DevOps Agent incident response hours, 10 DevOps Agent incident prevention hours, and 1000 chat messages per month.

Yes, you do need an AWS account to use AWS DevOps Agent.

If you are an administrator, log into the AWS Management Console within a designated AWS account that handles the agent's setup, access controls, and billing. Organize agent environments using "Agent Spaces," logical containers that define the scope and configuration for individual agent instances. Each Agent Space operates independently, allowing you to configure multi-account access, establish connections to third-party platforms (observability tools, pipelines, ticketing and collaboration systems), and manage organizational access permissions. Agent Spaces also serve as deployment points for custom integrations via remote MCP (Model Context Protocol) servers.

If you are an on-call engineer or an SRE who needs to work collaboratively with AWS DevOps Agent to resolve incidents and drive service improvements, you can interact with AWS DevOps Agent via a standalone AWS DevOps Agent web app. Via this web app, customers of AWS DevOps can view agent activity and findings and steer agent actions.

An Agent Space defines the scope of what AWS DevOps Agent can access as it performs tasks through IAM roles and tool integrations. You can create multiple Agent Spaces, typically aligned with team responsibilities or service boundaries, and extend them with integrations to observability tools, CI/CD pipelines, and incident management systems.

Yes, you can use AWS DevOps Agent with any existing or new AWS resources existing in one or many accounts.

No, AWS DevOps Agent does not use your content to train models.

AWS DevOps Agent currently supports English-language conversations.

AWS DevOps Agent uses Amazon Bedrock foundation models.

AWS DevOps Agent operates from the US East (N. Virginia) Region (us-east-1) as its primary processing hub. Support for additional Regions will be expanded before General Availability (GA). The system is designed to retrieve operational data from multiple AWS Regions across all AWS accounts that have been granted access within the configured Agent Space, enabling comprehensive visibility into distributed infrastructure and applications regardless of their geographic deployment. This multi-Region data collection capability ensures that the agent can perform thorough incident analysis and preventative recommendations across an organization's entire AWS footprint while maintaining centralized processing and control.

AWS DevOps Agent encrypts all customer data at rest using AES-256 encryption with AWS- managed keys, though Customer Managed Keys (CMK) are not supported until General Availability. Data is stored in us-east-1 with multi-Region replication available at General Availability.

AWS DevOps Agent maintains detailed journals that log every reasoning step and action taken, creating complete transparency into agent decision-making processes. Additionally, all agent activities are automatically captured by AWS CloudTrail within the hosting AWS account.