Google has expanded the capabilities of its AI-powered coding agent Jules, introducing a new command-line interface called Jules Tools and a public API to deepen its integration into developer workflows. The move aims to make Jules a more seamless, efficient part of software development by reducing context switching and enabling direct interaction within the terminal environment where developers already work. Jules Tools, now available, allows developers to initiate, monitor, and verify coding tasks directly in their terminal using simple commands. This shift from a chat-based interface to a CLI enables developers to stay within their primary workflow, delegating tasks like code generation, bug fixes, test writing, and performance improvements without leaving their current environment. According to Kathy Korevec, director of product at Google Labs, the goal is to minimize disruptions and let developers focus on building. Jules Tools is designed for scoped, self-contained tasks, contrasting with Google’s Gemini CLI, which supports more iterative and collaborative interactions. Jules operates asynchronously—once a task is approved, it formulates a plan and executes it independently, only pausing if it encounters an obstacle. This design reduces the need for constant back-and-forth, making it ideal for specific, well-defined coding challenges. In addition to the CLI, Google has opened the Jules API to developers, allowing them to embed the agent into custom systems, CI/CD pipelines, IDEs like VS Code, and even messaging platforms such as Slack. While the company is building native IDE plug-ins to enhance integration, the API gives teams flexibility to tailor Jules to their existing toolchains. This is especially valuable for organizations with established workflows and strong muscle memory around specific tools. Google has also enhanced Jules with new features, including memory functionality to remember user preferences, corrections, and feedback across sessions. Other recent updates include a stacked diff viewer layout, image uploads, and the ability to read and respond to pull request comments. These improvements aim to make Jules more context-aware and interactive. A key area of focus is reducing Jules’ dependency on GitHub. Currently, the agent operates within GitHub repositories, requiring users to either link an existing one or create a new blank repo. However, Google is exploring support for other version control systems and even environments without version control, responding to user demand for greater flexibility. Despite its growing capabilities, oversight remains important. Jules is designed to pause and ask for human input if it gets stuck, ensuring developers remain in control. However, mobile access is still limited—native notifications are not yet supported—though Google is actively improving the mobile experience, as many users access Jules via web browsers on phones. While Jules is primarily used by professional developers, some are experimenting with it as a complement to more casual coding platforms, taking projects beyond the limits of those tools and using Jules for advanced refinement. Launched in public preview in May and exited beta in August, Jules now operates under structured pricing. The free tier offers 15 daily tasks and three concurrent tasks. Paid plans—Google AI Pro at $19.99 and Ultra at $124.99 per month—offer up to five and twenty times the limits, respectively, catering to individual developers and teams with higher workloads. With these updates, Google is positioning Jules not just as a coding assistant but as a core, integrated component of modern software development. As AI reshapes how code is written, Jules’ evolution reflects a broader industry shift toward intelligent, autonomous collaboration in the developer ecosystem.