Managed Agents in the Gemini API
The Gemini API has introduced managed agents, allowing developers to create custom agents with specific instructions and skills. The new Antigravity agent operates in a secure cloud sandbox, enabling various functionalities such as reasoning and executing code. This development aims to simplify the infrastructure management for developers, focusing on enhancing product experiences.
- ▪Managed agents in the Gemini API allow for the creation of custom agents with user-defined instructions and skills.
- ▪The Antigravity agent operates in an isolated Linux environment, enabling reasoning, planning, and code execution.
- ▪Developers can define agents using markdown files, streamlining the process of agent management.
Opening excerpt (first ~120 words) tap to expand
Breadcrumb Innovation & AI Technology Developer tools Introducing Managed Agents in the Gemini API { "reading_time": "[[read\u002Dtime]] min read" } May 19, 2026 · Share x.com Facebook LinkedIn Mail Copy link The Gemini API now supports managed agents. Run the Antigravity agent in a secure cloud sandbox, build custom agents with your own instructions, skills and data and define them as versionable files using AGENTS.md and SKILL.md. Ali Çevik Member of the Technical Staff, Google DeepMind Philipp Schmid Member of the Technical Staff, Google DeepMind Share x.com Facebook LinkedIn Mail Copy link Sorry, your browser doesn't support embedded videos, but don't worry, you can download it and watch it with your favorite video player! Today, we're launching Managed Agents in the Gemini API.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Google.
Discussion
0 commentsMore from Google
