The landscape of software development is undergoing a massive shift as artificial intelligence transitions from a simple autocomplete tool to a sophisticated autonomous partner capable of understanding complex project architectures. In 2026, developers no longer settle for generic AI suggestions that lack context about their specific internal libraries or organizational coding standards. This demand for precision has led to the latest update for JetBrains IDEs, which introduces a suite of advanced features designed to bridge the gap between local development environments and enterprise-level AI orchestration. By integrating support for specialized agents and diverse model providers, the ecosystem now offers unparalleled flexibility for engineering teams. These enhancements ensure that the transition from conceptualization to deployment remains fluid, minimizing the friction often associated with switching between different AI tools. The focus has shifted from merely generating code to managing the entire development lifecycle through intelligent, context-aware assistance.
1. Enterprise Integration and Interactive CLI Management: Streamlining Team Workflows
Administrators now have the authority to provide standardized agents for their teams to use within the IDE, ensuring that every developer follows the same architectural patterns and security protocols. The process begins when enterprise or organization leaders create and publish custom agents for their members directly on GitHub. To utilize these specialized tools, developers must open the Chat panel in their JetBrains IDE and click on the selection tool to access the agent selector within Copilot Chat. From this menu, users can pick a company-wide or organization-specific agent that is best suited for their current project. After choosing the specific agent required for the task, the developer can begin working with the agent using the pre-set configurations provided by the organization. This centralized approach allows companies to maintain a high level of consistency across various projects, transforming the AI into a bespoke technical expert that understands the specific business goals and private codebases of the firm.
The latest update also introduces sophisticated real-time message management in the Copilot CLI, allowing users to interact with the system even while a request is still being processed. Developers can now utilize three new options to manage their workflow: they can place a message in line to queue a follow-up request, direct the current process to pivot as soon as a tool execution ends, or terminate and restart the session to cancel an ongoing response and send a new prompt. This level of granular control ensures that the AI remains responsive to the developer’s needs, even in the middle of a complex multi-step operation. Beyond these interaction controls, the update brings enhanced agent debugging through a new logs summary view. This is a consolidated dashboard in the Agent Debug panel that provides an overview of session activity and aggregate stats. These improvements to the CLI and debugging provide the transparency and control needed to trust AI tools in high-stakes environments, making the interaction feel more like a live collaboration.
2. Multi-Model Support and Platform Optimization: Enhancing Precision and Stability
Claude is now available as an alternative agent provider in a public preview, offering a powerful perspective on complex logic and multi-file reasoning tasks. To enable it, users must first set up the Claude Code CLI tool by downloading and installing the necessary binary on their computer. Following this installation, the user needs to configure the CLI file path in their IDE settings by navigating to the GitHub Copilot Chat settings and pointing the IDE to the specific Claude installation. Once configured, the developer can choose Claude from the available agents using the agent picker to start a session. This setup is complemented by model picker improvements that make navigating between different AI engines faster. Users can now use the /models shortcut to quickly open the selection menu, pick an expanded context window for data-heavy tasks, and access their most frequent models via the “recently used” section. These enhancements ensure that developers can apply the most appropriate tool to any specific problem they encounter in their daily work.
Transparency regarding resource consumption was improved with the introduction of a per-turn AI credits indicator, showing exactly how many credits were used for each interaction in Local, CLI, and Claude sessions. Alongside this, the update included refinements to the chat input design and inline chat state management to ensure a more reliable experience when handling complex prompts. Technical stability was a primary focus, as the team resolved redundant diff window openings and corrected the display of completion models. Interface hang issues were eliminated, ensuring the platform remained reliable during heavy processing. Finally, the release reached a major milestone as the Cloud Agent moved into general availability, meaning it no longer required a preview flag. Organizations were encouraged to audit their pipelines to identify where custom agents could reduce technical debt. The transition to enterprise-ready tools represented a significant advancement in how teams collaborated with AI. Developers utilized these capabilities to improve overall code quality.
