agentsclimarketplace

Cli first

Skill vibestackmd/vibestack/skills/cli-first

Convention for using CLI tools and environment variables when interacting with third-party services like AWS, Vercel, Supabase, Stripe, GitHub, and Google Cloud. Auto-loads when working with external services, deployments, infrastructure, or API integrations.From its SKILL.md

Install
npx -y skills add vibestackmd/vibestack --skill cli-first

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.1 KB, as published. Nobody here has run it

CLI-First Development

When interacting with third-party services, always prefer CLI tools over web dashboards, REST APIs, or connector plugins.

Why CLI First

  • CLI tools give you full control over a platform from the terminal, no clicking through dashboard menus
  • AI agents can orchestrate complex multi-step workflows (deploy, configure, monitor) through CLIs directly
  • CLIs are often more powerful and up-to-date than web UIs or third-party connectors
  • Everything stays scriptable, reproducible, and version-controllable

Environment Variables

Before making API calls or using SDKs directly, check .env* files for existing credentials and project configuration.

# Check for environment files in this order:
# 1. .env.local    , local overrides (gitignored, highest priority)
# 2. .env          , shared project defaults
# 3. .env.development / .env.production, environment-specific

Look for:

  • API keys and access tokens
  • Project IDs, org IDs, and region settings
  • Database connection strings
  • Service-specific configuration (bucket names, queue URLs, etc.)

Never hardcode credentials. If a needed credential isn't in .env*, ask the user to add it rather than creating one.

Common CLI Tools

When these CLIs are available, use them instead of raw API calls:

ServiceCLICommon Uses
AWSawsS3, Lambda, CloudFormation, IAM, ECR, ECS
VercelvercelDeploy, env vars, domains, project settings
SupabasesupabaseDB migrations, edge functions, auth config
GitHubghIssues, PRs, releases, Actions, repo settings
StripestripeWebhooks, test events, product/price setup
Google CloudgcloudCompute, Cloud Run, IAM, storage, pub/sub
FirebasefirebaseHosting, Firestore rules, functions
CloudflarewranglerWorkers, KV, R2, DNS

Workflow

  1. Check if the CLI is installed, run command -v <tool> or which <tool>
  2. Check .env* files for project credentials and configuration
  3. Check auth status, most CLIs have a whoami or status command
  4. Use the CLI to perform the operation instead of visiting the web dashboard
  5. If the CLI isn't installed, suggest the user install it (see extras/ for the dev tools installer) rather than working around it with raw HTTP calls

Examples

# Instead of visiting the Vercel dashboard to add an env var:
vercel env add MY_SECRET production

# Instead of clicking through AWS console to upload to S3:
aws s3 cp ./dist s3://my-bucket/ --recursive

# Instead of using the Supabase web UI to run a migration:
supabase db push

# Instead of manually creating a GitHub release:
gh release create v1.0.0 --generate-notes

# Instead of configuring Stripe webhooks in the dashboard:
stripe listen --forward-to localhost:3000/api/webhooks

Keep looking

Skills are one crate of 609,689. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.