Learn how subagents in Gemini CLI act as specialized experts to handle complex, high-volume tasks in isolated context windows. This new feature enables parallel execution, reduces context rot, and allows for custom agent definitions using simple Markdown and YAML.

GoogleDevs' platform is a central hub for developers interested in Google technologies, APIs, and developer tools. Through articles, tutorials, and documentation, GoogleDevs offers insights into building applications using Google Cloud Platform, Android, Chrome, and other Google services. Developers can learn about cloud computing, machine learning, and mobile app development with Google's developer tools and platforms.

Google Developers

Gemini CLI now supports subagents — specialized expert agents that operate in isolated context windows with their own tools, MCP servers, and system instructions. The main session acts as an orchestrator, delegating sub-tasks to the most relevant subagent. Custom subagents are defined via Markdown files with YAML frontmatter and can be stored globally or per-project. Parallel subagents allow multiple agents to run simultaneously for tasks like researching multiple topics or refactoring distinct components. Built-in subagents ship with Gemini CLI, and explicit delegation is possible using the @agent syntax.

Subagents have arrived in Gemini CLI

Build your own expert with custom subagents