Google Labs is expanding its AI programming assistant Jules with features that increase its practical application. These include Jules Tools, a command-line interface, and an API that allows the assistant to be integrated into existing systems and workflows.
Jules is designed as a digital programming colleague that can help generate code, fix bugs, write tests, and improve performance. Whereas the assistant has mainly worked via a chat interface until now, Jules Tools will enable developers to work with the AI directly in their terminal.
The API offers even more flexibility because Jules can be linked to existing development processes. For example, the assistant can automatically take action as soon as a bug is reported in Slack or when a build in a CI/CD pipeline gets stuck.
When Jules was introduced earlier this year, Google emphasized that the agent runs on Gemini 2.5 Pro and is capable of working asynchronously. This means that developers can set tasks, shut down their laptops, and retrieve the results later, without the need for continuous interaction.
Jules also integrates with GitHub and can clone repositories in a virtual cloud environment to perform independent operations on the code. During the beta, it became apparent that repetitive tasks, such as adjusting dependencies or generating test scripts, were frequently left to Jules. Google is thus positioning the tool as a broader addition to existing AI tools that typically work synchronously.
Developers want control and stability
Google points out that the introduction of Tools and API stems from requests from developers who want more control and stability. In recent weeks, the company has been working on reliability, lower latency, and resolving recurring issues with environment settings and file management. In addition, features have been added that make it more practical to use, such as a file selector for more targeted context, memory for user preferences, and a structured way to manage environment variables.
The enhancements demonstrate that Google Labs is positioning Jules not only as an experiment, but as a concrete tool for software teams. By no longer limiting the AI to a chat interface but making it available in terminal and via APIs, the step seems to have been taken towards broader application in serious development environments.