agent-platform-model-registry

What problem does it solve?

Manually managing machine learning models and their versions in the Agent Platform Model Registry is time-consuming and prone to human error, especially when handling frequent uploads, metadata updates, and deletions across multiple model versions.

Core Features & Use Cases

• Read-only Model Discovery: Quickly list all registered models and retrieve full metadata for specific models or versions without risk of accidental changes.
• Safe Model Upload & Update: Register new models or new versions of existing models, and update metadata like display names and labels, with built-in confirmation safeguards to prevent unintended modifications.
• Secure Model Deletion: Permanently remove unused models and all their versions with explicit confirmation steps to avoid irreversible data loss. Use case example: If your team maintains 20+ custom ML models for different Agent Platform use cases, use this skill to quickly audit all registered models, update outdated display names, and delete deprecated models without manual CLI command entry.

Quick Start

Use the agent-platform-model-registry skill to list all machine learning models currently registered in your Agent Platform Model Registry for your active Google Cloud project.