جوجل تطلق رسميًا أداة برمجة الذكاء الاصطناعي جولز بعد انتهاء مرحلة التجريب

Google has officially launched its AI coding agent, Jules, out of beta, marking a significant step in its evolution as a developer tool. Introduced initially as a Google Labs project in December and unveiled to the public during the I/O developer conference in May, Jules now joins the company’s suite of AI-powered tools with a stable, production-ready release. Powered by Gemini 2.5 Pro, Jules is an asynchronous AI agent that operates within Google Cloud virtual machines, enabling it to handle complex coding tasks independently. It integrates directly with GitHub, clones codebases, and autonomously fixes or updates code while developers continue working on other priorities. Unlike synchronous tools such as Cursor or Windsurf, which require real-time interaction, Jules allows users to initiate tasks and step away—returning hours later to find completed work, effectively acting as a virtual assistant. The decision to exit beta was driven by improved stability and extensive feedback gathered during the public preview. Over the past two months, Google implemented hundreds of UI and performance enhancements. Kathy Korevec, director of product at Google Labs, emphasized that the tool’s long-term viability is now well-established, citing strong user engagement and real-world usage patterns. With the general availability launch, Google introduced structured pricing tiers. The free "introductory access" plan allows 15 individual tasks per day and three concurrent tasks—down from the beta’s 60-task limit—designed to help users evaluate Jules on actual projects. Paid tiers are included in Google AI Pro ($19.99/month) and Google AI Ultra ($124.99/month), offering 5× and 20× higher task limits, respectively. Pricing reflects insights from beta usage, ensuring alignment with real developer workflows. Privacy remains a key focus. Google updated its policy to clarify that public repository data may be used for AI training, but private repositories are not shared with Google’s training systems. The change was prompted by user feedback on clarity, not a shift in data practices. During beta, thousands of developers completed tens of thousands of tasks, resulting in over 140,000 code improvements shared publicly. Feedback led to key enhancements: reusable project setups, GitHub issue integration, and support for multimodal inputs (e.g., text and images). A new GitHub integration now enables automatic pull request creation, while "Environment Snapshots" preserve dependencies and installation scripts for faster, consistent execution. User behavior revealed surprising trends: 45% of visits came from mobile devices, with India, the U.S., and Vietnam leading traffic. Though no dedicated mobile app exists, the web interface is widely used, prompting Google to explore mobile-specific features. Additionally, many users began with "vibe coding" tools—quick, experimental AI-generated projects—and used Jules to refine them for production. Initially requiring an existing codebase, Jules now supports empty repositories, broadening its accessibility to newcomers testing AI coding tools. Internally, Google already uses Jules on select projects, with plans to expand its adoption across the company. Jules represents a shift toward autonomous, scalable AI assistance in software development—offering developers not just code suggestions, but a reliable, hands-off collaborator.