Google Jules: AI Coding Agent Enters Developer Toolchains

Google Expands Jules AI Coding Agent Capabilities

Google is integrating its artificial intelligence coding assistant, Jules, more deeply into the workflows of software developers. This expansion includes a new command-line interface and a publicly accessible API. The aim is to allow Jules to function seamlessly within terminals, CI/CD systems, and communication platforms like Slack.

New Tools for Developers

Previously, access to Jules – Google’s asynchronous coding agent – was limited to its website and GitHub platform. The introduction of Jules Tools, a command-line interface (CLI), now provides developers with direct access to the agent from their terminal.

This CLI facilitates interaction with the agent through commands, optimizing workflows by removing the need to constantly switch between web interfaces and GitHub. Developers can remain within their preferred environment while delegating coding tasks and verifying the outcomes.

Reducing Context Switching

“Our primary goal is to minimize disruptions to the developer’s focus,” explained Kathy Korevec, Product Director at Google Labs.

Jules vs. Gemini CLI

Google also provides the Gemini CLI, an AI-powered command-line tool compatible with various developer environments, including terminals and CI/CD pipelines. Both tools leverage Google’s Gemini 2.5 Pro AI model. However, Korevec clarified to TechCrunch that Jules Tools is intended for “highly specific tasks,” while Gemini CLI necessitates a more “iterative” approach and greater user collaboration.

API Access and IDE Integration

In addition to the CLI, Google has released Jules’ API to the public, previously used for internal development. This allows developers to integrate Jules into their existing workflows and leverage their established expertise.

Developers can utilize the API to connect Jules with their integrated development environment (IDE), such as VS Code. Google’s team is actively working on developing dedicated plug-ins for IDEs to further enhance Jules’ integration capabilities.

Recent Enhancements

These updates follow the recent addition of “memory” functionality to Jules, enabling it to retain a history of user interactions, preferences, and feedback. The tool has also incorporated features like a stacked diff viewer layout, image uploading, and the ability to process and respond to pull request comments.

Expanding Code Hosting Options

Google is currently investigating ways to reduce Jules’ dependence on GitHub. Currently, the agent operates within a GitHub repository, requiring developers to connect it to an existing repository or create a new one.

“Users have expressed a desire for Jules to support other code hosting services,” Korevec stated. “We are exploring compatibility with various version control systems and options for users who do not require or prefer not to use version control.”

Ensuring Oversight and Mobile Accessibility

Maintaining oversight of AI tools is crucial, particularly in professional contexts. Jules is designed to alert the user if it encounters difficulties or reaches an impasse, requesting assistance.

“If the agent faces an issue or cannot resolve a problem independently, it will pause and ask for guidance,” Korevec explained.

However, oversight is more challenging on mobile devices due to the lack of native notification support. Google is working to improve the mobile experience, with a focus on implementing native notifications.

Target Audience and Pricing

Jules is primarily aimed at software engineers and other technical professionals, differentiating it from many “vibe coding” platforms geared towards non-programmers. However, some users are exploring Jules as a complement to more creative coding environments.

“We’ve observed users leveraging Jules to expand upon projects initiated in other coding tools when they reach their limitations,” Korevec shared.

Jules was launched in public preview in May, exited beta in August, and is now available under tiered pricing plans. A free plan provides access to up to 15 daily tasks and three concurrent tasks. The Google AI Pro and Ultra plans, priced at $19.99 and $124.99 per month respectively, offer approximately 5x and 20x the task limits.