Managed Agents in the Gemini API
Google launched Managed Agents in the Gemini API: one API call spins up an autonomous agent in a sandboxed Linux environment.
“you make a single API call and you get an autonomous agent that can work on behalf of you and solve like problems creatively.”
At Google I/O, the Gemini API team announced Managed Agents, where a single API call provisions an autonomous agent running in an isolated remote Linux sandbox that can write/execute code, run bash commands, use skills, and create files without touching the developer's own servers. Initially powered by Gemini 3.5 Flash and the Google anti-gravity agent harness (which also powers the anti-gravity IDE), it lets developers build custom agents with their own system instructions and skills, extending an agent strategy that began with the Interactions API and its first managed agent, Deep Research. It matters because it pushes hosted, server-side autonomous agents into mainstream API access, lowering the barrier to deploying agentic workloads.