terminal_input · Terminal Control MCP: Risk & Policy

WHAT IT DOES

What terminal_input does on Terminal Control MCP

AI agents invoke terminal_input to trigger actions in Terminal Control MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.

WHY IT NEEDS A POLICY

Why terminal_input needs a policy

This tool sends keyboard input to a live terminal session, which can trigger arbitrary commands, scripts, or interactions within TUI applications. The blast radius is high because an AI agent could use this to execute any shell command or interact with sensitive terminal-based applications by simulating keystrokes.

From the tool's definition 'Send input to a terminal session' combined with server description: 'simulating keyboard input through a virtual X11 display'

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

terminal_input Rate-limited

terminal_input stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "terminal_input": {
      "limits": [
        {
          "counter": "terminal_input_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Terminal Control MCP, so you can run agents against it safely.

FAQ

Questions about terminal_input

What does the terminal_input tool do? +

Send input to a terminal session. It is categorised as a Execute tool in the Terminal Control MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on terminal_input? +

Register the Terminal Control MCP server in PolicyLayer and add a rule for terminal_input: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Terminal Control MCP. Nothing to install.

What risk level is terminal_input? +

terminal_input is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit terminal_input? +

Yes. Add a rate_limit block to the terminal_input rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block terminal_input completely? +

Set action: deny in the PolicyLayer policy for terminal_input. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides terminal_input? +

terminal_input is provided by the Terminal Control MCP server (taskhub-sh/terminal-driver-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.