Use this skill to help deploy code to an environment. Triggered by "deploy", "deploy to staging", "deploy to production", "ship it".
Scanned 5/27/2026
Install via CLI
openskills install Imran-ml/claude-skills---
name: deploy
description: |
Use this skill to help deploy code to an environment. Triggered by
"deploy", "deploy to staging", "deploy to production", "ship it".
argument-hint: "[environment: staging|production|preview]"
allowed-tools: "Bash, Read"
disable-model-invocation: true
---
# Deployment Helper
Safely deploy to `$ARGUMENTS` (defaults to `staging` if not specified).
## Pre-Deploy Checklist
!`git status && echo "---" && git log --oneline -3`
### 1. Environment Confirmation
**Target:** $ARGUMENTS
⚠️ If deploying to **production**, I will explicitly confirm with you before running any deploy commands.
### 2. Pre-flight Checks
```bash
# Ensure clean working tree
git status
# Run tests
npm test
# Run linter
npm run lint
# Check for security issues
npm audit --audit-level=high
# Build
npm run build
```
### 3. Deploy Steps
Depending on your setup, deploy with one of:
```bash
# Vercel
vercel --prod # production
vercel # preview
# Railway
railway up
# Fly.io
fly deploy
# AWS (via CDK or SAM)
cdk deploy
sam deploy
# Kubernetes
kubectl apply -f k8s/ --namespace=production
# Docker Compose
docker-compose -f docker-compose.prod.yml up -d
# Heroku
git push heroku main
# GitHub Actions (trigger workflow)
gh workflow run deploy.yml --ref main -f environment=$ARGUMENTS
```
### 4. Post-Deploy Verification
```bash
# Check health endpoint
curl -f https://your-app.com/health
# Tail logs for errors
# vercel logs --follow
# fly logs
# kubectl logs -f deployment/app-name
```
### 5. Rollback (if needed)
```bash
# Vercel: revert to previous deployment in dashboard
# Fly.io: fly releases list; fly deploy --image <old-image>
# Kubernetes: kubectl rollout undo deployment/app-name
# Git: git revert HEAD && git push
```
## Rules
- Always run tests before deploying
- Always confirm before deploying to production
- Keep a rollback plan ready
- Monitor error rates for 10 minutes post-deploy
$ARGUMENTS
No comments yet. Be the first to comment!
Build reusable Terraform modules for AWS, Azure, and GCP infrastructure following infrastructure-as-code best practices. Use when creating infrastructure modules, standardizing cloud provisioning, or implementing reusable IaC components.