Superpowers
obra/superpowersA 15-skill development methodology that makes coding agents clarify the goal, write a plan, work through it with subagents and verify before calling it done.
Recommended
Check first Depth 3/5
google-agents-cli-workflow skill
The entry skill of agents-cli: a phased agent-development lifecycle from design dialogue through scaffold, build, eval, deploy, publish and observe.
Our verdict Recommended
A solid process skill with a clear phase table; worth installing as the entry point if you use agents-cli.
What the google-agents-cli-workflow skill does
The workflow skill is the always-active starting point of the agents-cli suite. It lays out phases from 0 to 7: understand the goal through a design dialogue, study recipes, scaffold a project, build with ADK, evaluate, deploy, publish, and observe. A table tells the agent which sibling skill to reload before each phase, because context compaction may have dropped earlier skill content. It insists that a scaffolded project exists before agent code is written, since skipping that loses eval boilerplate, CI/CD config and project conventions, and it requires a matching google-agents-cli version.
The skill adds code-preservation rules, model-selection guidance, troubleshooting, and a product-name mapping for older Google names (for example Vertex AI Agent Engine to Agent Runtime). At 22 KB it carries six reference files: brainstorming, commands, extension, internals, a spec template and terminology. Its description is written to trigger on general requests like developing, debugging or deploying an agent.
Good for
- Using an agent as the main developer of an ADK agent from idea to production
- Keeping a long agent session on a defined process
Not for
- Tasks unrelated to agents-cli
- Quick scripts that do not need a scaffolded project
Depth 4/5
A 22 KB SKILL.md with six reference files including a spec template; no scripts or evals.
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 entry-point skill that lays out the agent development lifecycle (scaffold, build, evaluate, deploy, publish, observe) and points to the sibling agents-cli skills. It tells the agent to install and run the agents-cli tool, which was not reviewed; extensions of that tool can run arbitrary code but prompt before install. The tool sends usage telemetry unless you opt out.
Worth knowing
- Relies on a CLI or package we could not reviewRequires the agents-cli binary (install: uv tool install google-agents-cli), whose code is not in the skill folder.SKILL.md:19
- Installs packages or toolsTells the agent to run
uv tool install google-agents-clioruvx google-agents-cli setup(the setup step installs the skills) if the tool is missing.SKILL.md:56 - Runs remote code at an unpinned versionFallback instruction
uvx google-agents-cli setupruns the PyPI package without a version pin (elsewhere the file pins ~=1.12.0).SKILL.md:37 - 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 (7 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 / uvx); third-party agents-cli extensions that references/extension.md says can be added (they run arbitrary code).
NVIDIA SkillSpector static scan: 5 pattern hits (Tool Misuse 2, MCP Rug Pull 2, Excessive Agency 1). Hits are documentation: '--skip-auth' is a flag for reinstalling skills, 'uvx google-agents-cli' is the install command, and 'run arbitrary code' is a warning in extension.md that non-first-party extensions prompt before install.
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. It structures a human-guided design dialogue and phased build, not an unattended run.
Alternatives
Mastra Agent Skills
mastra-ai/skillsMastra's official skills: a framework guide that routes to current docs and ten references, and a Factory supervisor skill for the mastra api CLI.
With caveats
Broad access Depth 3/5
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-workflow?
The entry skill of agents-cli: a phased agent-development lifecycle from design dialogue through scaffold, build, eval, deploy, publish and observe.
How do I install google-agents-cli-workflow?
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).
Why does the workflow skill refuse to write code first?
It requires a scaffolded project so that eval boilerplate, CI/CD config and conventions exist before agent code is written. It also asks clarifying questions in a design dialogue before scaffolding.
Sources: GitHub repository; install counts from skills.sh via Best Skills (CC BY 4.0). Review based on the published SKILL.md.