Data & Analysis ✓ Google · Official google-cloudagent-platformmodel-registrymodel-versioningmodel-metadatagcloud-cli

Agent Platform Model Registry Manager

Manage Agent Platform models, versions, and metadata from the gcloud CLI.

FollowSkills review · FSRS-2.0
Not recommended
48/ 100 5-point scale 2.4 / 5
Trust15 / 25 · 3.0/5

The skill explicitly separates read-only, reversible mutations, and destructive deletion, and requires tiered confirmation for uploads, updates, and deletion, with typed confirmation for deletion. This supports user control and transparency around external effects. Points are deducted because setup instructs direct authentication and project changes without least-privilege guidance, credential-handling or data-flow disclosure, dependency provenance, or rollback procedures; the prohibition on preflight checks conflicts with the later requirement that models be undeployed before deletion.

Reliability7 / 20 · 1.8/5

The main command paths, regional requirement, and operations for listing, describing, uploading, updating, and deleting are clearly documented with examples. Static evidence contains no tests, CI, compatibility guidance, abnormal-input handling, or diagnosable failure feedback; authentication failure, insufficient permissions, long-running uploads, invalid IDs, and missing variables are not addressed. The score is therefore limited under static review.

Adaptability9 / 15 · 3.0/5

The metadata clearly targets Agent Platform Model Registry management for models and versions, while excluding training, endpoint deployment, and non-Agent Platform models. Points are deducted because input requirements, output formats, semantic trigger boundaries, and permission or regional constraints are not defined. Chinese-language support and mainland-China network reachability are also unaddressed, while the core function depends on Google Cloud services.

Convention8 / 15 · 2.7/5

The document is organized into safety, environment setup, and operation sections, with usable command examples. The repository README supplies installation, indexing, Apache-2.0 licensing, issue reporting, contribution guidance, and an active-development notice. Points are deducted because the selected skill lacks its own version, changelog, named maintenance responsibility, dependency versions, FAQ, troubleshooting, and known-limitations section; repository-level governance does not fully substitute for skill-level maintenance information.

Effectiveness6 / 15 · 2.0/5

The skill covers the core registry actions and provides commands that can serve as direct operational starting points, including region and version targeting. Points are deducted because there are no verified success outputs, post-action validation steps, upload-parameter guidance, prerequisite permission details, or recovery workflows. Users must still supply important decisions for uploads, versioning, updates, and deletion, so static evidence supports only a limited effectiveness score.

Verifiability3 / 10 · 1.5/5

The source contains auditable commands, scope boundaries, and confirmation rules, while the README and LICENSE support verification of installation, licensing, and repository support paths. Points are deducted because there is no committed test suite, CI coverage, third-party execution evidence, changelog, or corroborating source material. This conclusion is based solely on static documentation and does not establish that the commands run successfully at the assessed revision.

Evidence confidence:Low Reviewed Jul 20, 2026 Reviewed revision 513a7a51e85f
The upstream repository has new commits since this review. The score still applies to the reviewed revision shown and may not cover the latest changes.
Before you use it
  • Before execution, verify Google Cloud credentials, project, region, and required IAM permissions, and avoid granting broader permissions than necessary.
  • The deletion workflow should validate deployment state and recoverability first; the current instructions both prohibit preflight checks and require undeployment, creating operational risk.
  • The skill does not explain handling of credentials, model artifacts, or sensitive metadata, and it lacks failure-recovery and result-validation procedures.
Review evidence [1][2][3][4]
See the full review method →

What does this skill do, and when should you use it?

This skill manages machine learning models and model versions in the Agent Platform Model Registry. It covers listing, describing, uploading, updating metadata, and deleting models. It is intended for registry operations, not model training, endpoint deployment, or non-Agent Platform models. Google Cloud authentication, application-default credentials, project configuration, and an explicit region are required before running commands.

Runs gcloud ai models list to discover registry models and numeric IDs; runs gcloud ai models describe for model or version metadata; runs gcloud ai models upload to register a custom model or a new version; runs gcloud ai models update to change display names, descriptions, or labels; and runs gcloud ai models delete to permanently remove a model and its versions. Read-only actions need no confirmation, uploads and updates require interactive confirmation containing the exact command, and deletion requires explicit typed confirmation.

  1. A Google Cloud user needs to discover existing Agent Platform registry models and their numeric IDs.
  2. A machine learning engineer needs the complete metadata for a model or a specific model version.
  3. A team needs to register a custom model and its artifacts in the Agent Platform Model Registry.
  4. A model owner needs to revise a model's display name, description, or labels.
  5. An administrator needs to permanently remove an obsolete model after its versions are undeployed from endpoints.

What are this skill's strengths and limitations?

Pros
  • Covers the core registry lifecycle: listing, describing, uploading, updating, and deleting.
  • Defines distinct confirmation requirements for read-only, reversible, and destructive actions.
  • Includes commands for inspecting versions and uploading new versions.
Limitations
  • Requires the Google Cloud CLI, Google Cloud authentication, and configured project and region.
  • Deletion is irreversible, and all model versions must be undeployed from every endpoint first.
  • The source provides no test suite, error-handling guidance, or troubleshooting procedure.
  • It does not cover model training, endpoint deployment, or non-Agent Platform models.

How do you install this skill?

Run npx skills add google/skills, then select Agent Platform Model Registry Management during installation. The README does not document the destination folder.

How do you use this skill?

First run gcloud auth login, gcloud auth application-default login, and gcloud config set project $PROJECT_ID. Always provide --region=$LOCATION_ID; for example, ask to list models in the Agent Platform Model Registry, which maps to gcloud ai models list --region=$LOCATION_ID. Complete the required confirmation flow before uploading, updating, or deleting.

FAQ

What setup is required?
The source requires Google Cloud login, Application Default Credentials, and an active project; it does not specify the required IAM roles.
Can uploads or updates run immediately?
No. They are Tier M operations and require interactive Yes/No confirmation containing the complete command and required flags, followed by approval in a later user turn.
Does deletion perform a deployment pre-check first?
The documented flow requests typed confirmation immediately rather than performing pre-flight checks first. All versions must already be undeployed from all endpoints.
Can this skill train models or deploy them to endpoints?
No. Its stated scope excludes model training, endpoint deployment, and management of non-Agent Platform models.

More skills from this repository

All from google/skills

Automation & Ops ✓ Google · Official

Agent Platform Model Deployment Assistant

Deploy, verify, and clean up Model Garden models on Google Agent Platform.

Automation & Ops ✓ Google · Official

Agent Platform Endpoint Management

Manage Agent Platform serving endpoints and troubleshoot common endpoint failures.

Dev & Engineering ✓ Google · Official

Agent Platform Prompt Manager

Manage Agent Platform managed prompts with Python SDK guidance.

Data & Analysis ✓ Google · Official

Agent Platform Model Tuning

A guided end-to-end workflow for fine-tuning open or Gemini models on Google Agent Platform.

Dev & Engineering ✓ Google · Official

Gemini Enterprise Skill Registry

Search, manage, monitor, and scaffold skills for the Gemini Enterprise Agent Platform.

Automation & Ops ✓ Google · Official

Google Cloud gcloud CLI Safety Skill

Helps agents manage and troubleshoot Google Cloud resources through validated, scoped gcloud commands.

Dev & Engineering ✓ Google · Official

Gemini API Enterprise Development Guide

Helps developers build enterprise Gemini applications on Agent Platform with Google's unified Gen AI SDK.

Data & Analysis ✓ Google · Official

AlloyDB Database Manager

Manage AlloyDB clusters, instances, and backups with CLI, IaC, client-library, and MCP guidance.

Dev & Engineering ✓ Google · Official

Agent Platform Tuning Manager

Inspect, list, and safely cancel Agent Platform GenAI tuning jobs.

Automation & Ops ✓ Google · Official

GKE Basics Navigator

A focused entry point for discovering GKE cluster needs and routing each task to the right specialist skill.

Automation & Ops ✓ Google · Official

Google Cloud Global Load Balancer Configurator

Design and deploy opinionated global external Application Load Balancers on Google Cloud.

Automation & Ops ✓ Google · Official

Google Cloud Developer Onboarding

Prepares an individual developer’s first Google Cloud project, billing setup, authentication, and initial deployment path.

Automation & Ops ✓ Google · Official

Google Cloud Foundation Builder

Deploys a secure enterprise landing-zone foundation for a Google Cloud organization.

Automation & Ops ✓ Google · Official

Google Cloud Storage Basics

Manage GCS buckets and objects while configuring access, protection, lifecycle, cost, and performance.

Automation & Ops ✓ Google · Official

Google Cloud AI Agent Builder

Design, implement, deploy, and validate AI agents and multi-agent systems on Google Cloud.

Data & Analysis ✓ Google · Official

Google Cloud Agent Evaluation Flywheel

Improve model and agent quality through structured evaluation, failure analysis, and iteration.

Dev & Engineering ✓ Google · Official

Google Agents CLI Agent Development Guide

Guides agents from specification through development, deployment, and monitoring.

Automation & Ops ✓ Google · Official

Gemini AI Studio to Agent Platform Migration

Guides Gemini API applications from Google AI Studio to Google Cloud Agent Platform.

Dev & Engineering ✓ Google · Official

Google Bigtable Basics

Design, configure, query, and troubleshoot Google Bigtable workloads.

Automation & Ops ✓ Google · Official

Google Cloud Live Multimodal Streaming Architect

Design and deploy Google Cloud solutions for live, bidirectional multimodal streams.

Related skills