send_keystrokes · Poof MCP: Risk & Policy

WHAT IT DOES

What send_keystrokes does on Poof

AI agents invoke send_keystrokes to trigger actions in Poof. 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 send_keystrokes needs a policy

Sending keystrokes to a terminal session can trigger arbitrary command execution (e.g., Enter to run a command, Ctrl+C to interrupt, etc.). The server context makes clear this is a terminal control interface, so keystrokes can have wide-ranging Execute-level effects depending on what is typed or submitted.

From the tool's definition send_keystrokes — sends key presses to a macOS Terminal.app session controlled via AppleScript; part of a server explicitly designed to 'control the macOS Terminal.app' and 'send keystrokes'

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

send_keystrokes Rate-limited

send_keystrokes 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": {
    "send_keystrokes": {
      "limits": [
        {
          "counter": "send_keystrokes_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 Poof, so you can run agents against it safely.

FAQ

Questions about send_keystrokes

What does the send_keystrokes tool do? +

Send key presses (e.g., [. It is categorised as a Execute tool in the Poof MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on send_keystrokes? +

Register the Poof MCP server in PolicyLayer and add a rule for send_keystrokes: 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 Poof. Nothing to install.

What risk level is send_keystrokes? +

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

Can I rate-limit send_keystrokes? +

Yes. Add a rate_limit block to the send_keystrokes 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 send_keystrokes completely? +

Set action: deny in the PolicyLayer policy for send_keystrokes. 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 send_keystrokes? +

send_keystrokes is provided by the Poof MCP server (mattapperson/poof-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.