Agents CLI in Agent Platform: Create to production in one CLI — Google Cloud collapses the local-to-production workflow into a single CLI that provisions infra and deploys agent services in hours instead of weeks. Outcome engineers can use this to standardize reproducible agent delivery pipelines and shorten the feedback loop between experiments and production (Principle 04/06).

Production-Ready AI Agents: 5 Lessons from Refactoring a Monolith — Google refactors a brittle sales-agent monolith into orchestrated ADK sub-agents with Pydantic outputs, dynamic RAG, and OpenTelemetry for reliability. The case study provides concrete orchestration, observability, and contract patterns you should replicate to make agent systems auditable and maintainable (Principle 09/14).

Agent Vault — Open-source credential proxy and vault for agents — Agent Vault brokers credentials via a local HTTPS proxy so agents never directly receive secrets, preventing credential exfiltration and enabling safer agent networking. Adding a credential proxy is a practical way to enforce least-privilege and reduce the biggest operational attack surface for agentic workflows (Principle 15/14).

Cyera acquires Ryft to give enterprises traceable data access for AI agents — Cyera merges Ryft’s data-lake tooling to provide traceable, governed data access for agents, tying automated retrieval to security controls. Traceable data paths materially improve grounding, auditability, and compliance for RAG-driven agents and should be part of your data-contract strategy (Principle 02/11).

85% of enterprises are running AI agents. Only 5% trust them enough to ship. — Cisco surfaces the enterprise trust gap and ships Defense Claw runtime controls and monitoring to close it. The statistic and tooling underscore that runtime governance, policy enforcement, and observability are gating factors for shipping agent-driven outcomes in production (Principle 14/15).