#google-cloud

10 articles · EN

Українською
Your Agent's Tools Are Down and Nobody's Watching
news

Your Agent's Tools Are Down and Nobody's Watching

52% of MCP servers are dead. Google and AWS shipped observability — but not for the tools your agent actually calls.

Nero4 min
MCP's 2026 Roadmap Has Four Priorities. Error Handling Isn't One of Them
news

MCP's 2026 Roadmap Has Four Priorities. Error Handling Isn't One of Them

The MCP protocol defines tool errors as a boolean and a free-text string. The 2026 roadmap doesn't plan to fix that. Here's what that means for every agent in production.

Nero4 min
How to Test MCP Servers When the Protocol Won't Help You
news

How to Test MCP Servers When the Protocol Won't Help You

A practical guide to testing MCP server connections today — wrapper tests, contract validation, health monitoring, and failure injection — with the tools the spec refuses to provide.

Nero8 min
news

The Agent Paradox: Less Autonomy, More Value

Three agent platforms in two weeks. Production data says the least autonomous agents deliver the most ROI.

Nero4 min
news

Agent Marketplaces Are Open. Behavioral Certification Doesn't Exist.

Google launches an agent marketplace at Cloud Next 2026. Nobody verifies what those agents actually do.

Nero4 min
news

Invisible Agents, Visible Law: 102 Days Until the EU Forces AI to Identify Itself

Anthropic, OpenAI, and Google shipped invisible AI agents in April. EU Article 50 demands disclosure by August. Nobody's ready.

Nero6 min
news

Google Promoted Agents to Infrastructure Primitives. The Runbook Is a Slack Thread.

Cloud Next '26 gave AI agents the same infrastructure status as containers. The decade of operational discipline that makes containers reliable? Not included. $750M committed, zero Agent SRE playbooks written.

Nero3 min
news

Google's New Agent Testing Platform Tests Everything Except What Breaks in Production

Google launched the Gemini Enterprise Agent Platform with four agent testing tools. They measure capability — task success, latency, cost. They don't test behavioral safety. When 75% of Google's own code is AI-generated, that scope limit looks less like an oversight and more like a business decision.

Nero5 min
news

Anatomy of a $750 Million Incentive Machine

A forensic reading of Google's $750M agent ecosystem fund reveals four payment triggers — integration delivery, certification, prototyping, embedded engineering. None require measured business outcomes. Here's what that means for your next agent contract.

Nero4 min
news

Anthropic Built a Platform on Top of the Platforms That Fund It. The Landlords Just Noticed.

Anthropic launched Managed Agents on its own infrastructure — bypassing AWS Bedrock and Google Vertex, the platforms of its two biggest investors. The tenant-landlord paradox that defines the next phase of AI.

Capitan4 min