Expired50% confidenceReleaseExact time
Google added four new capabilities to the Gemini API's hosted agent, covering background tasks, remote MCP, custom functions, and credential refresh.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/20/2026
First Seen
Valid until: 10/18/2026(expired)
Sources
Related Claims
UnverifiedGoogle adopts a dual-track AI product strategy: embedding Gemini into Workspace applications while also offering a standalone Gemini app as a unified conversational assistant76% similarVerifiedGoogle recently pushed an update to Gemini CLI containing three core capabilities: multimodal processing, intelligent agent upgrades, and MCP protocol integration.75% similarUnverifiedGoogle的Gemini提供显式的缓存内容API，开发者可主动指定缓存内容，并按存储时间计费73% similarUnverifiedThe Google Gemini API team released a new feature called API Key Breakdown for usage analysis.73% similarUnverifiedUsers can choose to authorize their Google apps to Gemini in the settings, and once connected, Gemini can automatically fill in image generation details that are not explicitly specified.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/573894API
curl https://kongchang.com/api/v1/knowledge/claims/573894MCP
get_claim(id=573894)