ClawHub
Back to skill

Security audit

Ma2 Control

Security checks for vulnerabilities and agentic risk

Overview

This skill openly controls a real grandMA2 lighting console, but it gives broad live and show-data-changing authority without clear confirmation safeguards.

Install only if you intentionally want an agent to control a grandMA2 console. Review the local ~/ma2_bridge scripts first, run them only on a trusted show network, verify the target IP and port, and require manual confirmation before any live output, playback, Store, Assign, Delete, Move, overwrite, noconfirm, or broad fixture-range command.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger conditions are broad enough that ordinary mentions of MA2, lighting consoles, executors, or cue operations could activate a skill that sends real control commands to a live grandMA2 console. In this context, unintended activation is materially dangerous because the skill is explicitly designed for state-changing operations over Telnet, so a false trigger can affect live show output or stored programming.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill enables live console-control actions such as selecting fixtures, changing intensity, storing cues, assigning executors, and triggering playback, but it does not prominently warn that these commands can immediately alter a running show or overwrite programming. Because the skill bridges to a real MA2 console over Telnet, the lack of strong safety messaging and confirmation increases the chance of destructive or hazardous operator error.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The document includes destructive MA2 commands such as copy/move/delete in a control skill that is specifically intended to drive a live grandMA2 console over Telnet. In this context, presenting destructive examples without safety notes, confirmation guidance, or production-vs-programming caveats can lead an agent or operator to issue state-changing commands that overwrite or delete cues during a live show or on valuable show data.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The template library includes state-changing grandMA2 commands such as Store Cue, Assign to Button/Executor, Go Executor, Goto Cue, and sequence assignment without any caution, confirmation requirement, or context limits. In a skill designed to send commands to a live lighting console over Telnet, these examples can normalize or encourage direct execution of commands that overwrite show data, alter playback bindings, or trigger live output, increasing the risk of accidental destructive or disruptive actions.

Static analysis

No suspicious patterns detected.