Google’s Jules AI Coding Agent: Now with Command Line Interface
Google has taken a significant step forward with its Jules AI coding agent. It has now been expanded with new command-line tools and a public API. This enhancement brings the autonomous coding assistant deeper into the developer workflows. The newly introduced Jules Tools command-line interface enables developers to interact with the AI agent directly from their terminal. This eliminates the need to switch between web interfaces and GitHub.
Jules, which is powered by Google’s Gemini 2.5 Pro model, operates asynchronously. It clones codebases into secure Google Cloud virtual machines and performs tasks such as bug fixes, writing tests, and code refactoring. Unlike synchronous AI coding tools like Cursor or Windsurf, Jules works independently. This allows developers to delegate tasks and return later to review the completed work.
“We want to reduce context switching for developers as much as possible,” said Kathy Korevec, director of product at Google Labs.
The public API enables integration with integrated development environments (IDEs) and CI/CD systems. Jules is available with a free tier offering 15 daily tasks. However, Google AI Pro and Ultra subscribers get 5x and 20x higher limits respectively.