Google Cloud Introduces MCP Server and Gemini to Simplify Management

Key Takeaways

1Google Cloud announces the general availability of the MCP server for Organization Policy

2Agents can audit, detect deviations, and manage exemptions without manual analysis

3Gemini allows users to write and test custom constraints in natural language

Why it matters : These tools facilitate the security and governance of cloud configurations, reducing risks associated with misconfigurations.

Google Cloud announces the general availability of the MCP server for its Organization Policy and the drafting of constraints in natural language with Gemini. The goal is to facilitate auditing, drift detection, and the application of guardrails, without resorting to complex CEL rules.

Dry-run validations and proactive governance

Gemini can generate both compliant and non-compliant test configurations to evaluate a policy and validate it in dry-run mode before its application. The Organization Policy ensures proactive governance of configurations on Google Cloud. It allows administrators to impose critical guardrails so that engineering teams can build and deploy quickly and securely. With a catalog of built-in guardrails, autonomous tools, and the customization of constraints across services, companies have the necessary control to evolve securely.

General availability of the MCP server and NL drafting

The general availability of the Organization Policy Model Context Protocol and the natural language constraint drafting powered by Gemini has been announced. The MCP server provides tools for autonomous agents and pipelines to interact directly with the Organization Policy, without manual analysis of complex hierarchies. These agents can assist administrators during posture audits, drift detection, and exemption management.

MCP tools: discover, assess, and manage guardrails

Constraint discovery is done through tools like ListConstraints and GetCustomConstraint, which provide information on the guardrails applied to a project, folder, or organization. GetEffectivePolicy calculates the effective policy considering hierarchical inheritance and tag-based conditions, and it can check if a configuration change will be compliant. For management, CreatePolicy and CreateCustomConstraint allow for the control of guardrails through conversational workflows.

Le brief IA que lisent les pros

Le brief IA que les pros lisent chaque soir

Les 7 actus IA du jour, décryptées en 5 min. Gratuit.

Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.

Choisissez votre rythme

Gratuit · Pas de spam · Désabonnement en 1 clic

Drafting custom constraints in natural language

The catalog provides managed constraints for common use cases, but each organization has specific architectural standards. The Custom Organization Policy allows these standards to be expressed as constraints and applied to supported services. Where rules were traditionally written in CEL, requiring mastery of Google Cloud resource schemas, administrators can now formulate them in natural language via a Gemini prompt. For example, they can request the use of the agent's identity in the Agent Runtime, impose an approved Agent Gateway for agent deployments, or prevent the disabling of automatic updates on GKE. Constraints can then be refined conversationally.

Use cases for agents and governance context

A coding agent can be associated with the MCP tools to set up a specific dashboard, display the active constraints, and identify folders and projects that are not subject to these rules. It is also possible to use custom-designed agents and skills to facilitate the administration of the Organization Policy. In a context where many incidents arise from avoidable misconfigurations, the Organization Policy Service helps secure resources within the Google Cloud hierarchy. It relies on over 100 predefined policies and the Custom Organization Policy to establish fine boundaries. Dedicated documentation is available to establish these fundamental guardrails.