computer_actions · Linux Computer Use MCP: Risk & Policy

WHAT IT DOES

What computer_actions does on Linux Computer Use

AI agents invoke computer_actions to trigger actions in Linux Computer Use. 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 computer_actions needs a policy

This tool executes arbitrary batches of desktop control actions on a Linux/X11 system. It can chain clicks, text input, keypresses, and scrolling in any combination, enabling full desktop automation. Misuse could trigger destructive OS-level operations (e.g., deleting files via file manager, running shell commands via terminal), making it critical severity.

From the tool's definition 'Batch multiple actions ({type:click|type_text|set_text|keypress|scroll, ...})' — combines clicking, typing, keypresses, and scrolling into arbitrary sequences of desktop control actions

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

computer_actions Rate-limited

computer_actions 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": {
    "computer_actions": {
      "limits": [
        {
          "counter": "computer_actions_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 Linux Computer Use, so you can run agents against it safely.

FAQ

Questions about computer_actions

What does the computer_actions tool do? +

Batch multiple actions ({type:click|type_text|set_text|keypress|scroll, ...}). It is categorised as a Execute tool in the Linux Computer Use MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on computer_actions? +

Register the Linux Computer Use MCP server in PolicyLayer and add a rule for computer_actions: 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 Linux Computer Use. Nothing to install.

What risk level is computer_actions? +

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

Can I rate-limit computer_actions? +

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

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

computer_actions is provided by the Linux Computer Use MCP server (tak-uukti/linux-computer-use). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.