keylog_stop

What keylog_stop does on Claude C2

AI agents invoke keylog_stop to trigger actions in Claude C2. 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 keylog_stop is rated High

This tool executes a command to stop a keylogger on a remote client system. While stopping a keylogger might seem benign, it is part of a suite of offensive/surveillance tools and represents execution control over remote system processes. The high severity reflects that an AI agent with access to this tool could disable security monitoring or forensic evidence collection during an attack.

From the tool's definition Tool name 'keylog_stop' and description 'Stop keylogger' indicate control of keylogging functionality. Combined with sibling tools like 'amsi_bypass', 'bypassuac', 'browser_creds', and 'defender_exclude', this server facilitates remote system compromise and…

The rule that runs keylog_stop safely

PolicyLayer is an MCP gateway: it sits between your AI agents and Claude C2, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For keylog_stop, this is the rule to start with:

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

Questions about keylog_stop