Research to Deploy

Research infrastructure best practices and generate deployment-ready cloud configurations.

Overview

This skill bridges the gap between researching cloud infrastructure patterns and actually deploying them. Instead of spending hours reading documentation, comparing approaches, and manually writing configuration files, this skill automates the entire pipeline: it searches for current best practices on the target platform, synthesizes the findings into a coherent deployment strategy, and generates production-grade Infrastructure as Code (IaC) that you can review and apply directly.

The skill supports multi-cloud deployments across GCP, AWS, and Azure, as well as platform-as-a-service providers like Railway, Fly.io, and Render. It generates Terraform modules by default but can also produce Pulumi programs, Docker Compose files, Kubernetes manifests, or platform-specific CLI commands. Every generated configuration includes security hardening, monitoring hooks, and cost optimization annotations based on the latest recommendations from the cloud provider.

Instructions

1. Describe what you want to deploy and where:

• "Research GCP Cloud Run best practices and deploy my Node.js API to staging"
• "Set up a production Kubernetes cluster on AWS with monitoring"
• "Create Terraform configs for a serverless Python function on Azure"

1. Specify constraints if you have them:

• Budget: "keep monthly costs under $50"
• Region: "deploy to us-central1"
• Compliance: "needs HIPAA-compliant storage"
• Existing infra: "we already use Terraform Cloud for state management"

1. Let the skill research. It will search for current documentation, community best practices, and known pitfalls for the specified platform and service. The research phase produces a summary of findings before generating any code.

1. Review the research summary and confirm the approach. The skill presents:

• Recommended architecture with rationale
• Cost estimate based on expected usage
• Security considerations and mitigations
• Alternative approaches that were considered

1. Apply the generated configs after review:

• Terraform: terraform init && terraform plan
• Docker: docker compose up -d
• Kubernetes: kubectl apply -f

Output

The skill produces a structured set of deployment artifacts:

• Research Summary (Markdown): A concise document covering the best practices found, architectural decisions made, and trade-offs considered. Includes source links.
• Infrastructure Code: Terraform modules (.tf files), Dockerfiles