Running DeepSource analysis locally has been one of the top requests from our existing and potential customers. We understand the importance of staying within your coding environment while performing code analysis and fixing issues. Today, we are excited to announce the private beta release of DeepSource for VS Code, which seamlessly integrates our powerful code health tools into Visual Studio Code.
But that’s not all — we've got something remarkable in store for you. All issues detected by DeepSource can now be automatically fixed right from the IDE, with our new Autofix engine backed by a powerful Large Language Model. No more dealing with hassles of manual fixes because all issues identified by DeepSource can be effortlessly resolved right from your IDE.
Seamless Analysis from the IDE
With support for all major programming languages, you can now effortlessly discover and address over 3000+ code maintainability and security issues without ever leaving your IDE. This will help you avoid back and forth in code reviews, as the IDE integration delivers instant feedback as you write, enabling you to identify and resolve potential issues before they are flagged as problems in pull requests. Best of all, every detected issue is accompanied by a detailed description, complete with examples and external references for further reading.
AI powered Autofix™
With our new LLM based Autofix engine, all issues identified by DeepSource can now be automatically fixed right from the IDE. Seamlessly integrated into the editor, this lets you accept or reject suggested fixes with a simple click, streamlining your workflow. By reducing the time spent on manual fixes and code review iterations, developers can focus on building new features and delivering value to end-users. Autofix is the first in a series of AI features in DeepSource. We are continuing our effort in training and fine-tuning custom models, to augment the capabilities of DeepSource, without compromising accuracy.
DeepSource for VS Code is now available in a private beta. If you would like to get early access, simply provide your information through this link, and we'll follow up via email to get you started. For teams interested in evaluating DeepSource, you can schedule a demo here.