Natural language query processor for advanced queries - AWS Config

Natural language query processor for advanced queries

The natural language query processor for advanced queries is in preview release for AWS Config and is subject to change.

The natural language query processor for advanced queries uses Amazon Bedrock, a generative artificial intelligence (generative AI) technology which allows you to ask questions in plain English and convert them into a ready-to-use query format. With the natural language query processor, you can query your AWS account or across an AWS organization.

You can provide your feedback about the generated query. Choose the thumbs up or thumbs down button that appears below the generated query.


Currently, the natural language query processor supports generative AI capabilities for the English language only.

Query using the natural language query processor

  1. Sign in to the AWS Management Console and open the AWS Config console at

  2. Choose Advanced queries from the left navigation to query your resource configurations for a single account and Region or for multiple accounts and Regions.

  3. On the Advanced queries page, choose New query to go to the Query editor.


    To run a query on an aggregator, create an aggregator. For more information, see Setting Up an Aggregator Using the Console.

    If you've already set up an aggregator, in the query scope, choose the aggregator to run an advanced query on that aggregator. When you select an aggregator, consider adding the AWS account ID and AWS Region in the query statement to view that information in the results.

  4. In the Query editor, go to the Natural language query processor section. Ask a question or make a request in plain English, and then choose Generate. For example, you can can ask one the following questions:

    • Which load balancers are created after January 1, 2023?

    • What are all publicly accessible RDS database instances?

    • Show the count of network ACLs that do not have any tags starting with key.

    • Provide a list of unencrypted Amazon EC2 volumes.

  5. After you generate a query, you can choose Copy (functionally equivalent to Ctrl+C) to copy the query to your clipboard, or choose Populate to editor. If populated to the editor, you can edit the generated query to fit your needs.

  6. To save this query to a new query, choose Save As.

    • In the Query Name field, update the name of the query.

    • In the Description field, update the description of the query.

    • Enter up to 50 unique tags for this query.

    • Choose Save.

  7. Choose Run. The query results display in the table below the query editor.

  8. Choose Export as to export the query results in CSV or JSON format.


    The query results are paginated. When you export, up to 500 results are exported.

    You can also use the APIs to retrieve all the results. The results are paginated and you can retrieve 100 results at a time.

Region Support

The natural language query processor is supported in the following Regions.

Region Name Region Endpoint Protocol
US East (N. Virginia) us-east-1 HTTPS
US West (Oregon) us-west-2 HTTPS