AWS Cloud Operations & Migrations Blog

Explore four new features in AWS Chatbot

AWS Chatbot helps you get started with ChatOps. For example, you can monitor AWS resources by receiving notifications in your Slack channels and interact with AWS resources by issuing diagnostic commands in your channels. In this post, I will share four new ways you can use AWS Chatbot to monitor and interact with AWS resources.

1. Amazon CloudWatch composite alarm notifications in chat channels

Amazon CloudWatch composite alarms combine multiple alarms to reduce noise and help you stay focused on critical operational issues. You can now receive composite alarm notifications in your AWS Chatbot configured channels.

These composite alarm notifications contain information about the parent alarm and up to three triggering alarm children in the ALARM state. Both the parent and children alarms contain links to the CloudWatch console where you can view alarm details.

To set up composite CloudWatch alarm notifications in AWS Chatbot, see Creating a Composite Alarm in the Amazon CloudWatch User Guide. In step 6 of the procedure, choose an Amazon Simple Notification Service (Amazon SNS) topic that you configured with AWS Chatbot. When the composite alarm is triggered, your chat channel will receive notifications.

CloudWatch composite alarm notification in AWS Chatbot says the alarm transitioned to the ALARM state because one of the data points was greater than the threshold. The alarm name is bill>750


Figure 1: Composite alarm notification example

2. Customize AWS CLI command output in AWS Chatbot

AWS Chatbot now supports the filtering of CLI output through an optional --query parameter. With this capability, AWS Chatbot users can customize the content of the CLI command output. Querying uses JMESPath syntax to create expressions for filtering your output.

For example, use the following to limit the cloudwatch describe-alarms CLI output to alarm name, description, state, and the reason attributes:

@AWS cloudwatch describe-alarms --query @.{MetricAlarms:MetricAlarms[*].{AlarmName:AlarmName, AlarmDescription:AlarmDescription, StateValue:StateValue, StateReason:StateReason}} --region us-east-2

For more information about CLI output filtering, see Client-side filtering in the AWS Command Line Interface User Guide and Running AWS CLI commands from Slack channels in the AWS Chatbot Administrator Guide.

3. Test Chatbot channel configuration

You can now use the AWS Chatbot console to test a chat channel configuration. Test messages are dispatched to all the Amazon SNS topics associated with the selected channel configuration and then delivered to the configured chat channels.

For more information, see Test notifications from AWS services to Amazon Chime or Slack in the AWS Chatbot Administrator Guide.

4. In-app feedback

At AWS, we pay attention to all customer feedback and use it to prioritize enhancements. You can now provide feedback directly from AWS Chatbot by typing the following command in your chat channel:

@AWS feedback I wish for AWS Chatbot to . . .

You can also use the Feedback button in the AWS Chatbot console.


In this blog post, I shared four new AWS Chatbot features that you can use to augment ChatOps in your organization. Give these features a try and tell us what you think by using the AWS Chatbot @aws feedback command. For more information about these features, see Using AWS Chatbot with other AWS services in the AWS Chatbot Administrator Guide.

About the author

Abhijit Barde

Abhijit Barde

Abhijit Barde is the Principal Product Manager for AWS Chatbot, where he focuses on making it easy for all AWS users to discover, monitor, and interact with AWS resources using conversational interfaces.