AI Tools & Products
3d ago
Gemini API Enhances Managed Agents with New Features for Developers
Jul 7, 2026
AI Summary
The Gemini API has introduced new features for Managed Agents, including background execution, remote MCP server integration, and custom function calling. These enhancements aim to improve the development experience by allowing for more reliable and efficient agent interactions.
- New capabilities for Managed Agents in the Gemini API include background execution, remote MCP server integration, custom function calling, and credential refresh.
- These updates respond to developer feedback and are designed to facilitate the creation of production-ready agents.
- Managed agents in the Gemini Interactions API operate through a single endpoint, handling reasoning, code execution, and file management within a secure cloud environment.
- Background execution allows for asynchronous interactions, enabling long-running tasks without maintaining an open HTTP connection.
- Developers can now connect managed agents directly to remote Model Context Protocol (MCP) servers, simplifying access to private databases and internal APIs.
- The API supports the addition of custom tools alongside built-in sandbox tools for local execution, enhancing flexibility in agent capabilities.
- Credential management has been improved, allowing for the refresh and rotation of access tokens and API keys without losing the sandbox's state.
- These updates position managed agents as asynchronous workers that can operate effectively within development environments.
gemini apimanaged agentsdeveloper toolsproduction-readybackground tasks