In brief: Gemini CLI now supports specialized subagents that handle complex tasks in parallel. Each subagent has its own tools and context window, making the main session faster and more efficient.
Google has introduced subagents for Gemini CLI. These specialized expert agents enable complex tasks to be automatically distributed across multiple independent systems, thereby increasing efficiency.
Subagents are specialized expert agents that work in parallel with the main Gemini CLI session. For complex or large-scale tasks, Gemini CLI acts as a strategic orchestrator and distributes individual subtasks to the most suitable subagents. Each subagent operates independently with its own tools, MCP servers, system instructions, and separate context window. This specialization ensures that the primary session remains fast and focused, while specialized agents handle intermediate tasks in parallel. Subagents make it possible to process complex, repetitive, or large-scale tasks more efficiently while keeping sight of the user’s overall vision.