Amazon Q Developer now has three new capabilities. Use /review to initiate a code review. Use /doc to ask Q Developer to build a knowledge graph of your code base, and then generate a detailed README for your project. Use /test to generate unit tests or improve existing unit test coverage.

Custom transformations (new and only in CLI): The CLI will allow you to perform custom transformations you defined specific to your code bases in your organization. Prior to this launch, Amazon Q Developer would upgrade open-source libraries in your Java applications. With custom transformations in the CLI, you can define your own transformations specific to your code bases and internal libraries. You can define custom transformations using ast-grep, a code tool for structural search and replace. Amazon Q Developer can perform your custom transformations and leverage Q’s AI debugging capabilities.

Amazon Q Developer Java upgrade transformation now offers step-by-step upgrades, and library upgrades for Java 17 applications. This new feature allows developers to review and accept code changes in multiple diffs, and to test proposed changes in each diff step-by-step. Additionally, Amazon Q can now upgrade libraries for applications already on Java 17, enabling continuous maintenance. Learn more from the documentation:

Eclipse IDE users can now install the Amazon Q Developer plugin. With this launch, developers can now chat with Amazon Q Developer about their project and code faster with inline code suggestions. You can also customize Q Developer to reference internal repositories to make coding suggestions more relevant to your organization. This plugin is currently in public preview.  

In the AWS console you can now Amazon Q more complex questions about your AWS bill. For example: 

• “What were my top 5 most expensive services last month?”
• “How much did my S3 costs increase between Q1 and Q2?”
• “Did we receive any credits last quarter, and if so, how much?”

Amazon Q Developer chat customizations allow you to securely connect Q Developer to your private codebases. You will then receive more precise chat responses that take into account your organization’s internal APIs, libraries, classes, and methods. Readmes and best practices demonstrated within your code repositories are also utilized within your customization. You can use a customized version of Q Developer chat in the IDE to ask questions about how your internal codebase is structured, and where and how certain functions or libraries are used.

Linux users can now access the same powerful autocomplete engine previously accessible only for macOS. This enables IDE-like autocomplete for hundreds of the most popular command line tools directly in the user's terminal. Try it out by downloading the desktop application, opening your preferred terminal, and running your favorite command line tools today.

Amazon Q Developer in the AWS Management Console now provides context-aware assistance for your questions about resources in your account. This feature allows you to ask questions directly related to the console page you're viewing, eliminating the need to specify the service or resource in your query. For example, while in the EC2 console try “what is the ami of this instance?”

Initially when you generated a code diff with Q Developer Agent you had to either accept the entire diff, or reject the entire diff with feedback on how to change it. You can now accept and apply individual files from the diff, or reject individual files and then apply the rest of the diff in one go.

DataDog is an observability and security platform for cloud applications. You can now ask questions like: "@datadog do I have any active alerts?"

Wiz is a cloud security platform that is used to discover security risks and track mitigation tickets. You can now ask questions like "What are my critical issues in Wiz?" and get immediate answers without leaving the AWS console.

Q Developer now features an intelligent, agentic workflow for code transformations that significantly improves Java modernization success rates. The system maintains full build context visibility across iterations, implements git-based state management, and leverages cross-iteration memory for consistent fixes. This substantial advancement in transformation capabilities delivers measurable improvements, particularly for Java 8/11 to 17 migrations. Try modernizing your Java applications today using /transform in your IDE.

Q will provide suggestions for all YAML files and all JSON “configuration” files. Previously, Q only provided suggestions for CloudFormation YAML/JSON files. Try it out inside a Node.js package.json file, or inside of a Kubernetes manifest.

Q will now provide longer, multi-line suggestions for Ruby, Scala, and SQL; a frequently requested enhancement. Try it out by typing a comment with an explanation of a longer function or code block that you’d like generated.

The new model adds inline suggestions support for the Dart, Lua, R, Swift, SystemVerilog, and PowerShell languages, along with the Vue.js framework (React is already supported). This brings the total number of languages/frameworks supported by Q inline suggestions to 26.

Both management and member account administrators can enable detailed per-user activity reports in the Amazon Q Developer settings by specifying an Amazon S3 bucket where the reports should be published. When enabled, you will receive a daily report in Amazon S3 with detailed user activity metrics, such as the number of messages sent, and AI lines of code generated.

You can now use Amazon Q inline in your IDE, without ever touching the mouse or using copy and paste. Press ⌘+I (Ctrl+I on Windows) to trigger inline chat. Describe a function or feature you’d like to develop and Amazon Q will generate and display a code diff that inserts new code at the cursor position. Press Enter to accept and apply the diff, or Escape to reject it. Alternatively you select a block of code (maybe even the entire file) then press ⌘+I (Ctrl+I on Windows) to provide instructions on how to refactor the selected code. You will see a diff against the selected code and can press Enter to accept and apply the diff.

When you start a shell interface in the AWS console, you now have automatic access to Amazon Q powered autocomplete suggestions, natural-language-to-code translation, and inline documentation.

You can now record your actions in the EC2, VPC, and RDS console, and automatically generate the CLI commands, Cloud Development Kit app, or CloudFormation template to recreate what you did. This new feature will assist with automation and the adoption of infrastructure as code.

When chatting with Amazon Q about your AWS resources, you can now click on links to jump directly to the referenced resource or resource list. Try it out by opening Amazon Q in the AWS console and asking a question like “what EC2 instances are running?”

Amazon Q is now allowed to generate longer code snippets (>20 lines of code) when chatting in the AWS console chat and AWS documentation. This update is accompanied by code references. When generated code appears similar to open source training data, you will now be able to see the attributed reference.  

When chatting with your workspace you can now ask questions that reference specific files. For example: @workspace what does factory.ts do? To try it out use the @workspace keyword in the IDE chat prompt, while also including a specific file name from your project.

For software development organizations that require auditing, debugging, analytics, or compliance, you can now enable prompt logging. On each chat or inline code completion, both the outbound prompt content as well as the response content that was returned will be logged into an S3 bucket. To query this data, consider using Amazon Athena.  

Previously, Q Developer Agent required human intervention to review and approve a work plan for how to write the code. We have made a significant internal improvement to how Q Developer Agent functions. It now combines the planning and coding phase, and works in iterations. This allows it to adapt on the fly and autonomously solve problems without direct human guidance. You will see increased accuracy and quality of generated code. Update to the latest version of the Q Developer IDE extension and use /dev to try it out.

As an administrator of Amazon Q Developer you can now see additional status about subscription status and subscribed applications, when viewing users in the Amazon Q Developer console.

Amazon Q Developer can help you troubleshoot errors from Amazon Elastic Compute Cloud (Amazon EC2), Amazon Elastic Container Service (Amazon ECS), Amazon Simple Storage Service (Amazon S3), and AWS Lambda. Generating these error remediation recommendations is now approximately 4x faster.

When you use /dev to describe a feature you want to develop, Amazon Q Developer creates a plan, then a code diff that delivers on that plan. We have made a significant internal improvement to Amazon Q Developer to improve how closely the code diff adheres to the plan. You should now see more complete code diffs, with fewer missed files.

You can now receive security scan findings for your Kotlin, Scala, and JSX code, through the Amazon Q Developer IDE extension. We have launched 26 Kotlin, 38 Scala, and 161 JSX rules. 

Previously Amazon Q Developer chat in your IDE was aware of your currently open file, but did not consider all project files. You can now add @workspace to chat messages. Amazon Q Developer will then consider all code files in the current project. Try out questions such as “@workspace what does this codebase do?” or “where does this @workspace implement authentication and authorization?”.

When you open the Amazon Q chat panel within the AWS console, you can now query resources on your AWS account with statements such as “list the EC2 instances in this region” or “show me the open ports on security group X”

Connect Amazon Q Developer to your organization's private code repositories to create customizations that will modify how inline code recommendation and chat function. Amazon Q Developer can now learn your org's coding style, write code snippets that utilize your private internal libraries, and better understand the overall structure of existing internal codebases.

Chat with Amazon Q, generate code, and debug error messages inside of Amazon SageMaker Studio. As a data science or ML engineer you will benefit from Amazon Q assistance throughout the machine learning development process.

Amazon Q is now generally available for Visual Studio 2022. Use Amazon Q Chat to help author, explain, refactor, fix, and optimize code. Run security scans on C# code to catch security vulnerabilities.

Amazon Q Developer Agent now has increased understanding of the following programming languages: C, C++, C#, Java, JavaScript, and TypeScript. We have made internal improvements to the syntax parser that allows Q Developer Agent to understand large codebases. This will enable improved code diffs that stick to the feature development plan.

Amazon Q Developer is now available in the AWS Console, and as part of the AWS Toolkit installed in Visual Studio Code, and JetBrains IDE’s.

Amazon CodeWhisperer, a real-time AI coding companion, is generally available and also includes a CodeWhisperer Individual tier that’s free to use for all developers.