What does this skill do, and when should you use it?
This skill guides agents working with Google Cloud Storage (GCS), an object-storage service built around buckets and immutable objects. It covers bucket and object operations through gcloud storage, gsutil, JSON/XML APIs, client libraries, Terraform, and Cloud Storage MCP servers. It also addresses IAM, ACLs, signed URLs, storage classes, lifecycle management, encryption, retention, soft delete, and performance. It is explicitly out of scope for Persistent Disk, BigQuery, and databases such as Cloud SQL, Spanner, Bigtable, and Firestore.
Enables the Cloud Storage API, creates buckets, and uploads, downloads, reads, lists, and transfers objects. It provides guidance for managing storage through the CLI, JSON/XML APIs, client libraries, Terraform, or MCP. It also covers access control, signed URLs, storage classes and tiering, lifecycle and cost management, versioning, CMEK, retention, object holds, soft delete, static website hosting, Pub/Sub notifications, gcsfuse mounts, and high-performance storage options.
- A developer needs to create a Google Cloud bucket and upload or download files with gcloud or an API.
- A cloud engineer needs to choose storage classes and locations for backups, website content, data lakes, or AI/ML training data.
- An administrator needs to protect objects with IAM, ACLs, signed URLs, encryption, or Bucket Lock.
- An engineering team needs to automate Cloud Storage management with Terraform, client libraries, or MCP.
- An operations user needs lifecycle rules, data transfer, Pub/Sub events, or gcsfuse mounts.
How do you install this skill?
- Before executing creation, update, deletion, IAM, public-access, transfer, or mount commands, confirm the target project, bucket, objects, identity, cost, and rollback plan.
- Do not expose access tokens, HMAC keys, cloud-provider credentials, or signed URLs through users, logs, or third-party services; prefer short-lived credentials and workload identity.
- The skill depends on Google Cloud endpoints, gcloud, and related services; mainland-China reachability, quotas, regional availability, and product availability are not verified by the source files.
- Validate version-sensitive commands and features against current official documentation and the locally installed tool versions.
- Shell / CLI
- Network access
- Local filesystem
- MCP Server
Google Cloud CLI (gcloud)curl
Run npx skills add google/skills, then select Google Cloud Storage Basics from the interactive prompt. The README does not document a specific installation directory or non-interactive selection arguments.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/google/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/cloud/google-cloud-storage-basics" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Using the Google Cloud Storage Basics skill, create a bucket in us-central1 and upload ./my-file.txt.
After installation, use a prompt such as “Using the Google Cloud Storage Basics skill, create a bucket in us-central1 and upload ./my-file.txt.” Real Cloud Storage actions require Google Cloud authentication and suitable permissions. The skill provides examples with gcloud, curl, and the JSON API, and recommends Cloud Storage MCP tools when an MCP server is connected.
What are this skill's strengths and limitations?
- Covers bucket and object CRUD plus security, lifecycle, transfer, and performance topics.
- Provides guidance across CLI, API, client-library, Terraform, and MCP workflows.
- Clearly distinguishes object storage from Persistent Disk, BigQuery, and database services.
- Includes practical operations topics such as cost optimization, data protection, and event notifications.
- The source provides no test suite, platform matrix, or validation results.
- Many workflows depend on Google Cloud authentication, IAM permissions, network access, and local CLI tools.
- The README documents interactive skill selection but not more detailed installation configuration.
- The skill focuses on Cloud Storage and does not cover block storage, data warehousing, or databases.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Google Cloud Storage Basics this page ✓ Google · Official | 56 · Use with care | ★ 21k | 12d ago | Apache-2.0 |
| Google Cloud Global Load Balancer Configurator ✓ Google · Official | 45 · Not recommended | ★ 21k | 12d ago | Apache-2.0 |
| Azure Deployment Runner ✓ Microsoft · Official | 54 · Use with care | ★ 2.7k | 3mo ago | MIT |
| Google Cloud gcloud CLI Safety Skill ✓ Google · Official | 51 · Use with care | ★ 21k | 12d ago | Apache-2.0 |
| Agent Platform Endpoint Management ✓ Google · Official | 50 · Use with care | ★ 21k | 12d ago | Apache-2.0 |
The skill covers gcloud storage/gsutil, JSON/XML APIs, Google Cloud client libraries, Terraform, and Cloud Storage MCP. When MCP is connected, it recommends structured MCP tools; otherwise, it presents gcloud storage and the JSON API as fallbacks.
How did FollowSkills review this skill?
The document discloses authentication methods, IAM, least-privilege roles, auditing, encryption, data protection, and attribution tagging; bucket deletion, recursive deletion, and irreversible retention locking require explicit confirmation or prominent warnings. Deducted 8 points because many commands directly mutate cloud resources without a uniform preflight-confirmation or rollback workflow, sensitive data flows are not fully described, and examples involve tokens, HMAC, external-cloud credentials, and remote services.
The skill provides CLI, JSON API, client-library, transfer, and mount paths, with several limitations and failure cases documented. Deducted 12 points because this is a static review with no reproducible tests or CI evidence; the broad, version-sensitive coverage provides limited diagnostics for invalid input, authorization failures, network failures, and unavailable commands.
The audience and scenarios are clear for GCS object storage, with CLI, API, libraries, Terraform, MCP, and FUSE coverage and explicit exclusions for block storage, data warehousing, and databases. Deducted 5 points because trigger conditions and input/output boundaries are not formalized, Chinese-language guidance is absent, and the core Google Cloud/overseas endpoints have no mainland-China reachability assessment.
The skill uses a layered quick start and reference directory, with installation notes, limitations, examples, cost guidance, security guidance, and references; the supplied metadata identifies Apache-2.0 and official Google provenance. Deducted 6 points because maintainer ownership, changelog, stable version policy, and update path are not clearly stated in the skill, while dependency and feature-version details are fragmented and incomplete.
Core bucket and object operations have directly reusable commands and code, with useful security, cost, transfer, and performance guidance. Deducted 8 points because static review cannot verify execution success; the scope is very broad, and many advanced paths require additional configuration, permissions, version alignment, or human review before producing a usable result.
The skill links extensively to Google primary documentation and includes auditable commands and configuration examples. Deducted 5 points because it contains no committed test suite, CI results, execution logs, or independent corroboration; static source alone cannot establish that the examples and version-sensitive claims run successfully.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →