Terraform MCP Server

What the Terraform MCP server does

How models use it and what it is built for.

The Terraform MCP server is a documentation and registry tool first. An agent searches for a provider, resolves it to a concrete version, and pulls the actual documentation for a resource or data source before it writes a single line of HCL — which is the difference between generated configuration that plans cleanly and configuration that fails on an unknown argument. The same search-then-fetch pattern covers registry modules and Sentinel policies. Beyond the public registry, setting a TFE address and token unlocks the HCP Terraform and Terraform Enterprise APIs, so an agent can list organizations and workspaces, inspect runs and read state versions. It runs over stdio for local clients and can also be started in streamable-HTTP mode when you want to host it for a team.

Tools the Terraform MCP server exposes

Typical tools an AI model can call. Exact names vary by version.

• search_providers — find a provider in the registry and resolve a documentation ID
• get_provider_details — the real documentation for a resource or data source at a version
• search_modules / get_module_details — registry modules, their inputs and outputs
• get_latest_provider_version / get_latest_module_version — pin against what actually exists
• search_policies / get_policy_details — Sentinel policy libraries from the registry
• HCP Terraform tools — organizations, workspaces, runs and state versions with a TFE token

Connecting to Terraform

Taken from the official Terraform documentation — see hashicorp/terraform-mcp-server — official repository for the full reference.

{
  "mcpServers": {
    "terraform": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "hashicorp/terraform-mcp-server"]
    }
  }
}

{
  "mcpServers": {
    "terraform": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "TFE_ADDRESS", "-e", "TFE_TOKEN",
        "hashicorp/terraform-mcp-server"
      ],
      "env": {
        "TFE_ADDRESS": "https://app.terraform.io",
        "TFE_TOKEN": "YOUR_TOKEN"
      }
    }
  }
}

Example prompts to try

Copy any of these into MCP Agent Studio after connecting.

• What are the required arguments for aws_ecs_service in the latest AWS provider?

• Find a registry module for a VPC with private subnets and show me its inputs.

• Write a Terraform config for a Cloudflare DNS record, checking the real docs first.

• Which of my HCP Terraform workspaces have runs that errored this week?

Models on MCP Playground

This is not a single-model product: you get the same MCP connection with 60+ models (Claude, GPT, Gemini, DeepSeek, open-weight, and more), you can switch mid-conversation, and you can open Compare mode to run the same prompt against multiple models at once. The card above is a suggested starting point for this server — not the only choice.

Check an AI agent can actually use the Terraform MCP server

Listing tools proves the server is reachable, not that a model can work with it. Evals go further: they read every tool on the server, write a test suite from its real schemas, and run it — code decides pass/fail on the responses (schema conformance, error codes, pagination, result caps) while a scoring model grades plain-English tasks driven through the tools.

Terraform MCP server — FAQ

Common questions about connecting, scoping and using it safely.