JetBrains Unveils 'JetBrains Air', an Integrated Product System for Agentic Development

JetBrains has unveiled 'JetBrains Air', a new product system for agentic software development. This marks a strategic shift from the traditional IDE-centric development workbench to a broader system encompassing the initiation, execution, orchestration, review, and governance of agent tasks.

Multi-vendor Environment and Governance

JetBrains believes the future of AI development lies in a Multi-vendor environment that is not dependent on specific models or vendors. However, mixing multiple AI services leads to visibility issues such as context fragmentation, inability to attribute costs, and the need to rebuild policies. JetBrains Air addresses this fragmentation by providing a single execution, visibility, governance, and accounting platform for all models and agents, without being tied to any specific model provider.

Key Components

JetBrains Air consists of the following core components:

• Air in JetBrains IDEs: Directs and orchestrates agents within the IDE, verifying the accuracy of agent outputs through JetBrains' code intelligence.
• Air Teams: Orchestrates and automates software delivery workflows involving developers and autonomous agents.
• Air Governance: Ensures policy enforcement, visibility, auditability, cost management, and accountability for AI-assisted development at the organizational level.
• Junie: JetBrains' coding agent specialized for software development, which is expected to be supported across all Air surfaces.

Agent Client Protocol (ACP) and Technical Foundation

JetBrains has introduced the Agent Client Protocol (ACP) to standardize the connection between IDEs and agent harnesses (planning, logic, tools, model routing, observability). This enables the discovery and execution of compatible agents, while JetBrains' 26 years of accumulated deterministic code intelligence technology serves as the foundation for enhancing agent reliability and efficiency. This approach allows agents to re-retrieve existing knowledge within the codebase, reducing reconstruction costs and improving accuracy.

Goals and Philosophy

The ultimate goal of JetBrains Air is not to maximize code generation volume. Instead, it aims to build a system that allows developers and organizations to scale experimentation while maintaining the ability to understand, verify, and stand behind the software. Since improvements in AI performance cannot replace organizational policies, provenance tracking, cost visibility, and accountability decisions, the focus remains on strengthening human control and organizational governance.