google-cloud_skill

Overview

This skill provides concise, practical guidance and reference material for working with Google Cloud across AI/ML, authentication, buildpacks, enterprise setup, generative AI, security, and Terraform. It centralizes common operational patterns, best practices, and troubleshooting steps to accelerate development, deployment, and governance on Google Cloud. Use it to find actionable next steps, configuration tips, and design tradeoffs for production systems.

How this skill works

I inspect documentation topics relevant to your question (AI/ML model selection and generative AI design, API keys and Application Default Credentials, Buildpacks and Procfile usage, enterprise foundation or Terraform deployment flows, security controls like BeyondProd and encryption, plus quotas and observability). I summarize recommended workflows, highlight configuration commands and options, and point out security and compliance considerations. When specific steps are needed I provide targeted, step-by-step guidance or configuration snippets you can apply immediately.

When to use it

• Designing or deploying generative AI applications and picking models or SDKs
• Setting up authentication, API keys, or Application Default Credentials for services and CLIs
• Preparing production-ready builds with Buildpacks or configuring Procfiles
• Provisioning enterprise foundations or deploying infrastructure with Terraform
• Implementing security controls: BeyondProd, encryption at rest/in transit, Cloud HSM, access revocation
• Troubleshooting quota limits, monitoring, or billing/export configurations

Best practices

• Use Application Default Credentials for server-side code and restrict API keys with IP or referrer rules
• Avoid embedding API keys in source; rotate and delete unused keys regularly
• Follow BeyondProd principles for runtime enforcement and supply-chain protections (Binary Authorization, signed artifacts)
• Use Buildpacks for reproducible images and include a Procfile for process definitions when needed
• Deploy enterprise foundation with a reproducible Terraform flow and keep state in Cloud Storage with proper IAM
• Monitor quotas and set alerts; request quota increases proactively for planned rollouts

Example use cases

• Build a generative AI prototype using Google Cloud SDKs, choose a model based on latency and cost targets
• Migrate application authentication to ADC and configure gcloud for CI/CD pipelines
• Package a web app with Buildpacks, include a Procfile, and deploy to Cloud Run or GKE
• Bootstrap an organization with Terraform templates provided in the setup flow and manage provisioning in Cloud Shell
• Harden production services by enabling Binary Authorization, CMEK, and Cloud HSM for key protection