Google is significantly enhancing its AI coding agent, Jules, by introducing a new command-line interface (CLI) and making its API publicly available. These strategic updates aim to embed Jules deeper into existing developer workflows, allowing seamless integration with terminals, continuous integration/continuous deployment (CI/CD) systems, and popular communication tools like Slack. This move intensifies the competition among tech giants striving to define the future of software development as an increasingly AI-assisted endeavor.

Previously, Jules, Google’s asynchronous coding agent, was primarily accessible through its dedicated website and GitHub. The introduction of Jules Tools, a new command-line interface, now brings the agent directly into the developer's terminal. This CLI enables developers to interact with Jules using simple commands, thereby streamlining their workflows and significantly reducing the need for constant context switching between the web interface, GitHub, and their development environment. As Kathy Korevec, director of product at Google Labs, emphasized, the primary goal is to minimize context switching for developers, allowing them to remain within their familiar environment while delegating coding tasks and validating results efficiently.

Google already offers another AI-based command-line tool, Gemini CLI, which operates across various developer environments. Both Jules Tools and Gemini CLI are powered by Google’s Gemini 2.5 Pro AI model. However, Korevec clarified that Jules Tools is specifically designed for "very scoped tasks," suggesting a more autonomous operation. In contrast, Gemini CLI encourages users to be "a lot more iterative" and engage in greater collaboration with the tool. Denise Kwan, Google’s senior developer advocate, further elaborated that Jules is designed to be less interactive, executing tasks independently once the user has approved its initial plan, thus offering a distinct approach compared to Gemini CLI.

Beyond the CLI, Google has also released Jules’ API to the public, an interface previously used exclusively for internal development. This public API empowers developers to extend Jules’ capabilities into their existing workflows, leveraging their "muscle memory and familiarity." The API facilitates integration with various integrated development environments (IDEs), such as VS Code, by allowing developers to build custom connections. Google Labs is also actively exploring the creation of specific plug-ins for popular IDEs to further solidify Jules’ presence and utility within the development ecosystem.

These latest updates build upon several recent enhancements to Jules. A notable addition is the implementation of "memory," which allows the agent to maintain a record of past interactions, user preferences, nudges, and corrections, fostering a more personalized and efficient experience. Over the past few weeks, Jules has also gained other valuable features, including a stacked layout for its diff viewer, the ability to upload images, and the functionality to read and respond to comments on pull requests, collectively enriching its utility for developers.

Google is also looking ahead to reduce Jules' current reliance on GitHub. Currently, the agent operates within a GitHub repository, necessitating either a connection to an existing repo or the provision of a new, blank one. Korevec indicated that users desire integration with other code hosting providers and that Google is exploring ways to support various version control systems, or even scenarios where a version control system is not preferred or necessary. This aims to provide greater flexibility and broader applicability for Jules across diverse development practices.

Addressing the critical aspect of oversight in AI tools, especially in professional contexts, Jules is designed with built-in safeguards. The agent is programmed to notify the user if it encounters an issue or gets "stuck" on a particular task, prompting the developer to intervene and assist. Korevec explained that if the tool cannot resolve a problem independently, it will pause and pose a question to the user. While many users are already accessing Jules through its mobile web interface, Google is actively working to enhance the mobile experience, particularly by developing native notifications to address the challenges of oversight on smaller screens.

Jules has primarily found its user base among professional software engineers and other technical professionals, distinguishing it from many "vibe coding" platforms often aimed at non-coders. However, some users are leveraging Jules to complement more casual or creative coding environments, bringing projects that have reached their limits in simpler tools to Jules for advanced extension and development. Launched in public preview in May and exiting beta in August, Jules is now available under structured pricing tiers, including a free plan offering up to 15 individual daily tasks and three concurrent tasks. Higher usage limits are available through the Google AI Pro and Ultra plans, priced at $19.99 and $124.99 per month respectively, providing approximately 5x and 20x the limits of the free tier.