HCP Terraform MCP server

What the model can do

HCP Terraform (Terraform Cloud) and Terraform Enterprise: organisations, projects, workspaces with version, lock and resource count, runs with status, message and timing, run comments, and the names of a workspace's variables (never their values or state). Can comment on a run, queue a plan or a run, apply, discard or cancel a run, and lock or unlock a workspace.

Reads

• terraformcloud_list_organizationsOrganisations the HCP Terraform token can see.
• terraformcloud_list_projectsProjects in an HCP Terraform organisation, which group workspaces: id, name and description.
• terraformcloud_list_workspacesWorkspaces in an HCP Terraform organisation: name, project, Terraform version, execution mode, auto-apply, whether locked, resource count, VCS repo…
• terraformcloud_list_runsRuns, newest first: status (applied, errored, planned, policy_checked, discarded…), message, workspace, source, trigger…
• terraformcloud_get_runOne HCP Terraform run in full: status and every status timestamp, message, workspace, source, trigger, whether it has changes…
• terraformcloud_list_variablesWhich variables a workspace has: key, category (terraform or env), whether sensitive or HCL, and description.

Acts, when you ask

These change something in HCP Terraform. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• terraformcloud_add_run_commentAdd a comment to an HCP Terraform run, recording a decision or feedback on it.
• terraformcloud_create_runQueues a run on an HCP Terraform workspace with its latest configuration: a plan, then an apply.
• terraformcloud_apply_runConfirms and applies an HCP Terraform run that is waiting for confirmation after its plan, changing real infrastructure as the plan says.
• terraformcloud_discard_runDiscards an HCP Terraform run that is waiting for confirmation or priority, so it is never applied and later runs can proceed.
• terraformcloud_cancel_runCancels an HCP Terraform run that is planning or applying, like ctrl+c: Terraform stops as safely as it can…
• terraformcloud_lock_workspaceLocks an HCP Terraform workspace so no run can start in it until unlocked, for everyone.
• terraformcloud_unlock_workspaceUnlocks an HCP Terraform workspace so runs can start again.

Connect it

First connect HCP Terraform on Dashboard → Connections, then add Memtro to your client once. The HCP Terraform tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared HCP Terraform connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Why did the last run on the networking workspace error?
• › Who applied last on production, and when?
• › Which workspaces are locked right now?
• › Is AWS_REGION set on the billing workspace?
• › Apply the pending run on the Northwind database workspace

Questions

Is this an official HCP Terraform MCP server?

No. It is Memtro's MCP server with Memtro's HCP Terraform connector behind it: the connector reads HCP Terraform through its API with the credentials you connect on Dashboard → Connections, and HCP Terraform publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in HCP Terraform?

Only when you ask, and only what it has been granted: add a comment to a Terraform run; queue a Terraform run (a plan, and an apply unless plan-only) on a workspace; apply a Terraform run waiting for confirmation; discard a Terraform run waiting for confirmation; cancel a Terraform run that is planning or applying; lock a Terraform workspace; unlock a Terraform workspace. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Engineering and operations

• GitHub MCP server
• AWS CloudWatch MCP server
• Sentry MCP server
• ServiceNow MCP server
• Okta MCP server
• Twilio MCP server