Gemini API gains background execution & MCP support for agents
Google DeepMind has updated Gemini API's managed agents with support for background execution, Model Context Protocol (MCP), and custom functions. These enhancements aim to increase flexibility for AI-driven automation in enterprise solutions.
Vad har hänt
On July 8, 2026, Google DeepMind introduced significant updates to its managed agents within the Gemini API. The new features including support for background execution, integration with the Model Context Protocol (MCP), and the ability to utilize custom functions. Additionally, the platform has been enhanced to handle seamless credential updates without loss of state.
Key facts
| Funktioner tillagda | Bakgrundskörning, MCP-stöd, anpassade funktioner |
|---|
Varför det spelar roll
The updates are intended to make Gemini API agents more adaptable for business applications requiring long-running processes and secure, dynamic workflows. Background execution enables asynchronous task management, which is valuable for system monitoring and data synchronisation. MCP integration facilitates better connectivity to external tools and services.
Vem påverkas
Developers building applications with the Gemini API are directly affected through expanded possibilities for creating more robust and autonomous AI agents. Enterprises using AI-driven automation can benefit from increased flexibility and efficiency in their systems. Users of AI services may indirectly experience improved performance and reliability.
EU-status
Ej relevant för EU-status.
Mer att veta
The ability to use custom functions alongside existing sandboxed tools gives developers greater control over agent behaviour and interaction with the external environment.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vem påverkas av uppdateringarna?
Länken öppnar i nytt fönster och leder till utgivarens egen sida.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.