TestingCatalog reports Google Gemini Business now supports custom MCP server connections so agents can attach private data and internal tools. The post also asks how soon consumer Gemini will get the same.
SpaceXAI engineer Baconbrix said Grok now auto-suggests connectors based on what it needs for better results. Agent context updates to include the new tools without a restart, and connections persist across chats β useful for continuous tool use in AI coding workflows.
Per Ars Technica, cloud models from OpenAI, Anthropic, xAI, and Google saw rare overlapping disruptions over several hours Thursday morning ET. Anthropic reported elevated errors on Claude Mythos 5.1, Fable 5.1, and Opus 5 at 9:23 ET and resolved by about 12:16 ET; OpenAI reported degraded ChatGPT and Codex performance from 10:43 ET, mitigated by about 12:55 ET. xAIβs Grok showed a user-facing outage banner as DownDetector reports jumped from under 10 to about 1,365; Google did not publicly acknowledge issues, but DownDetector and StatusGator pointed to a likely Gemini API outage around 10:45β11:15 ET. AWS, Azure, and Cloudflare had no major incidents. Concurrent frontier-model downtime is almost unheard of.
An AWS Machine Learning blog covers hosting coding agents on Amazon Bedrock AgentCore Runtime: each session gets a Firecracker microVM, /mnt/workspace survives idle, agentcore exec --it provides a real PTY, and Gateway MCP keeps GitHub/Jira tokens out of the VM. Run Claude Code, Codex, Kiro, and Cursor in isolation without localhost port fights.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.