Cloudflare Skills
cloudflare/skillsCloudflare's official 19-skill pack for building on Workers, Durable Objects, the Agents SDK, Cloudflare One and more, with an optional bundled MCP server.
Recommended
Check first Depth 4/5
google-agents-cli-deploy skill
Guides an agent through deploying ADK and other agents to Agent Runtime, Cloud Run or GKE, with Terraform, CI/CD, secrets and rollback.
Our verdict Recommended with caveats
Detailed and practical, but it drives real cloud deployments, so use a non-production project first and read Terraform plans.
What the google-agents-cli-deploy skill does
The deploy skill covers getting an agent from a scaffolded project to production on Google Cloud. A decision matrix compares three targets: Agent Runtime in the Gemini platform (managed scaling, sessions handled by the platform), Cloud Run (configurable instances, concurrency and networking) and GKE (full Kubernetes control). It prefers the agents-cli commands that wrap Terraform, Docker and the deployment steps, and covers service accounts, secrets, rollback and troubleshooting common failures such as 403 errors and timeouts.
At 31 KB it is the largest skill in the set, with eight reference files: Agent Runtime, Cloud Run, GKE, Terraform patterns, batch inference triggers, the full CI/CD pipeline with workload identity federation, testing a deployed agent with curl and load tests, and customer-managed encryption keys. It points to the observability skill for tracing and to the publish skill for registration with Gemini. The real actions are performed by the agents-cli binary against your cloud project.
Good for
- Taking an ADK agent to Cloud Run, GKE or Agent Runtime
- Setting up CI/CD and Terraform for an agent project
Not for
- Deploying to non-Google clouds
- Local-only prototypes
Depth 4/5
A 31 KB SKILL.md with eight reference files covering targets, Terraform, CI/CD and testing; no scripts or tests in the folder.
Dependencies and costs
- agents-cli (uv tool install google-agents-cli), Python 3.11+, uv and Node.js
- A Google Cloud project and credentials for deployment and cloud features (an AI Studio API key is enough for local create, run and eval)
- Telemetry: The agents-cli tool sends one usage signal per command (command name, exit code, error class, duration, CLI and Python versions, OS), written as a log entry to your own Google Cloud project's Cloud Logging, never file contents or arguments. Opt out with AGENTS_CLI_TELEMETRY=0; agents-cli info shows the current state.
Security review
Review before installingMarkdown-only guide to deploying agents to Agent Runtime, Cloud Run or GKE with agents-cli, including Terraform and CI/CD setup. The commands it describes create cloud resources, service accounts and IAM bindings in your Google Cloud project using your credentials; the skill says deploy must have explicit human approval. The agents-cli tool was not reviewed and sends usage telemetry unless you opt out.
Worth knowing
- Relies on a CLI or package we could not reviewRequires the agents-cli binary (plus gcloud and terraform in the instructions), none of whose code is in the skill folder.SKILL.md:20
- Installs packages or toolsInstall step is
uv tool install google-agents-cli.SKILL.md:25 - Makes network requestsShows curl calls to the user's own deployed service URL (and a local GKE port-forward) to test the agent, and gcloud/Terraform calls to Google Cloud.references/testing-deployed-agents.md:82
- Sends usage data by default (can be turned off)The agents-cli tool this skill drives sends per-command usage telemetry by default; the repo README (outside the skill folder) documents AGENTS_CLI_TELEMETRY=0 and DO_NOT_TRACK=1 to opt out.README.md:195
What we checked
The skill folder (9 files) on Oct 29, 2026. Not a full review: it relies on code outside the repository: agents-cli (PyPI package google-agents-cli, installed with uv tool install); gcloud CLI and Terraform invoked by the documented commands.
NVIDIA SkillSpector static scan: 17 pattern hits (Privilege Escalation 6, analysis-evasion 5, Tool Misuse 2, Excessive Agency 2, Data Exfiltration 2). Hits are documentation: 'analysis-evasion partial' notes refer to ordinary reference markdown, credential matches are Secret Manager/CMEK instructions, curl commands are tests against the user's own service. '--no-confirm-project' skips a project-selection prompt, but the same file says never to run deploy without explicit human approval.
A review, not a certification. Skills change after we check them — read the files before installing.
License
Apache-2.0 — commercial use allowed.
Can it run unattended?
Not really. Deployments change billable cloud resources and need human review of permissions and plans.
Alternatives
Azure Skills Plugin
microsoft/azure-skillsMicrosoft's official plugin: 94 Azure skills plus the Azure MCP Server (200+ tools) and Foundry MCP for preparing, deploying, diagnosing and costing Azure workloads.
Recommended
Broad access Depth 4/5
Google Agent Skills
google/skillsGoogle's official collection of about 180 skills for Google Cloud, Gemini APIs, Analytics, Data Manager, sign-in and other Google products.
Recommended
Broad access Depth 3/5
FAQ
What is google-agents-cli-deploy?
Guides an agent through deploying ADK and other agents to Agent Runtime, Cloud Run or GKE, with Terraform, CI/CD, secrets and rollback.
How do I install google-agents-cli-deploy?
It ships in the google/agents-cli repository. Run uvx google-agents-cli setup or npx skills add google/agents-cli; the skill expects the agents-cli binary (uv tool install google-agents-cli).
Is google-agents-cli-deploy safe to run?
The skill itself is only markdown, but it directs the agent to deploy to your Google Cloud project, which creates billable resources and IAM changes. Use a test project and review each plan before applying.
Sources: GitHub repository; install counts from skills.sh via Best Skills (CC BY 4.0). Review based on the published SKILL.md.