On demand
Claude on Google Cloud: Monitoring and Securing Agents at Scale
Autonomous agents can now take actions across systems, make decisions without manual human review, and operate at scale. That creates real questions for security and IT teams: how do you enforce what an agent is allowed to do, and how do you trace and audit what it actually did. Join Anthropic and Google Cloud for a technical session on answering both in production — building Claude agents with Claude Agent SDK, applying guardrails at every tool call, and sending traces and audit events into Google Cloud Observability.
Featured speakers
Roy Arsan
Applied AI @ Anthropic
Ivan Nardini
Developer Relations Engineer (AI/ML) at Google Cloud
What you’ll learn
How to give each agent a scoped identity and enforce per-agent and per-user policies with Agent Identity, Cloud IAM, and Agent Gateway
How Model Armor is designed to detect and block prompt injection, tool poisoning, and data leakage at runtime
How Anthropic's Agent SDK emits OpenTelemetry traces and audit events into Cloud Observability, and how its hooks and permission layers let you enforce policy on every tool call