Terraform MCP server (tfmcp)

This is a helper program called tfmcp that connects your AI assistant to Terraform, the tool many teams use to set up and manage cloud infrastructure. Once it is connected, you can ask your AI in plain words to check your Terraform files, show what a change would do, and even apply it. It is handy for anyone who works with cloud setups but does not want to memorize every Terraform command.

What this MCP server does

You ask your AI something like "what would this change do?" The AI sends that request to tfmcp, which runs Terraform inside your project folder. tfmcp reads your configuration, prepares the run, creates a plan, and shows what would be added, changed or removed. It can also review the plan for risks, apply it after you approve, and check the result. Everything happens locally on your machine, using the Terraform you already have installed.

Click to zoom

What you can do with it

• Read and analyze your Terraform configuration files
• Create a plan and explain what it would change
• Review a plan for risks, blockers and destructive changes
• Apply a reviewed plan and check the result
• Inspect the current state and look for drift
• Format your Terraform code and show a dependency graph
• Look up providers, modules and policies in the Terraform registry

Try asking your AI

• “Look at my Terraform project in ~/work/infra and tell me what it sets up”
• “Create a plan for this project and explain what would change”
• “Review the plan and warn me about anything risky or destructive”
• “Apply the plan we just reviewed and confirm it worked”

What it gives back to you

You get back plain answers in the chat: a summary of your configuration, a list of what a plan would add, change or remove, and any warnings or blockers. For a review, it gives you a risk assessment with recommendations. After an apply, it tells you whether it succeeded, the exit code, and whether the state was verified.

Before you start

Who it's for

Developers, DevOps engineers and cloud teams who already use Terraform and want their AI assistant to help read, plan and apply changes.