Safety #Google#AI agents Nov 22, 2026 · Originally published Nov 20

Google previews Semantic governance policies: an LLM judge checks every tool call an agent proposes against user intent and plain-language rules before it runs

Permissions say what an agent can do, not whether a particular call follows your rules. How Google puts that check outside the agent, and which model APIs it works with.

Image: Google Developers Blog
Primary source Google Developers Blog · developers.googleblog.com Read the original ↗

// Key points

  • Available in preview in Gemini platform and enforced through Agent Gateway: all agent egress goes through the gateway, and when the model returns a tool call, the policy engine evaluates it before the response reaches the agent.
  • Each call must pass two checks: your rules written in plain language (such as "before the earnings release, quarterly results may be shared only with @example.com addresses"), and a user-intent check, on by default, that catches calls drifting from the original request, as with indirect prompt injection.
  • If a call fails, the engine replaces the response content, tool calls included, with a configurable denial message so the call never runs; every verdict, the judge's rationale and the tokens used go to Cloud Logging. Text-only responses are not evaluated.
  • Agents can call their model through the Gemini API, the Anthropic Messages API, Claude models in Gemini platform, or OpenAI-compatible Chat Completions APIs, including self-hosted servers such as vLLM.

Builder's takeThe point is where the check sits: rules in a system prompt can be argued away by a user or a document, a check at the gateway cannot. Even off Google's platform you can copy the shape: put an independent judge in front of write or send tools, at least checking that the call still matches what the user actually asked, and log every verdict and rationale. For a backend like PandaClaws that publishes content automatically, I'd try this layer on the outbound tools first.

// Background · from #Google

Full timeline →
  1. Nov 20 Gemini API previews Tool Retrieval: tool parameter schemas load on demand instead of sending the whole tool library on every request
  2. Nov 20 The Gemini app for Mac gets computer use: it completes multi-step tasks across local files, desktop apps and the web in an isolated background window, with key actions needing your approval
  3. Nov 20 The Genie 4 world model comes to the Gemini API: interactive 720p worlds at 24 fps with synchronized native audio, from a text prompt or reference image

Every Monday, the week in AI in one email

Every Monday morning: the 6 AI stories that mattered most in the past seven days, with my take, in your inbox.

Your email is only used to send the weekly and is never shared. Every email has a one-click unsubscribe link. See the latest issue →