Skip to main content

Automation with Claude

Terraform and GitHub Automation with Claude

Terraform step by step, the mistakes that burn teams, and a GitHub workflow where Claude reviews and a human approves.

0:00 / 0:00
  • Instructor-led training: Online or Destination
  • 3 days · 24 hours
  • Starts in about 3 weeks
  • Flexible and customizable

What you will be able to do

  • Write, plan and apply Terraform configurations step by step with providers, variables, modules, remote state and locking
  • Recognise and fix the common Terraform mistakes: unpinned providers, secrets in code, state in Git, count versus for_each, targeted applies and drift
  • Run a GitHub repository for infrastructure code with branches, pull requests, CODEOWNERS, branch protection and GitHub Actions plan-on-PR and apply-on-merge
  • Set up Claude Code, the Claude Agent SDK and an MCP server so Claude can draft, plan and review Terraform without ever running apply itself
  • Build a plan-review agent that reads terraform show -json output and posts a structured risk verdict to the pull request
  • Enforce guardrails, environment approvals and an audit trail for every plan and apply across F5, NetScaler, Cisco ACI and other load balancers (Anaya LB)

Table of contents

  1. 1Infrastructure as code for network and load balancer teams
  2. 2Setting up Claude Code, the Agent SDK and MCP for Terraform work
  3. 3Terraform scripting step by step: init, plan, apply, state
  4. 4Common Terraform mistakes and how to catch them early
  5. 5GitHub repository fundamentals for infrastructure code
  6. 6GitHub Actions pipelines: plan on pull request, apply on merge
  7. 7Guardrails, approvals and audit for Terraform and GitHub automation
  8. 8Claude as plan reviewer: Agent SDK, MCP and structured verdicts
  9. 9Terraform for F5, NetScaler, Cisco ACI and other load balancers
  10. 10Capstone: end-to-end change through the Claude-reviewed pipeline