ClawHub
Back to skill

Security audit

Home Music

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to be a home music automation tool, but it needs review because its local command execution and broad trigger phrases could cause unintended or redirected playback automation.

Review the skill before installing. Use it only if you intentionally want an agent to control local Spotify and speaker routing, prefer explicit invocations such as naming the skill, and check that HOME_MUSIC_SCRIPT and PATH cannot redirect it to an unexpected executable.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
return 0

    try:
        result = subprocess.run(cmd, check=False)
        if result.returncode != 0:
            print(f"[home_music] scene '{scene}' exited with code {result.returncode}")
            if raise_on_error:
Confidence
89% confidence
Finding
The code executes an external program using a path that can be overridden by the HOME_MUSIC_SCRIPT environment variable or resolved from PATH, with no validation that the target is the expected trusted script. While subprocess.run is invoked without shell=True, reducing shell-injection risk, this still enables arbitrary code execution if an attacker can influence the environment or PATH seen by the calling automation.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger phrase "house music" is a common natural-language request and is likely to collide with ordinary user intent rather than a deliberate invocation of this skill. In a voice or chat-routing system, that can cause accidental activation of local automation that controls Spotify and speaker routing without clear user intent.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The example voice phrases are everyday commands like "start party mode" and "stop the music," which are ambiguous and can be spoken in many contexts. If an agent uses examples to derive trigger behavior, these broad phrases can lead to unintentional execution of speaker and playback controls, especially for disruptive actions like enabling loud playback or stopping media.

Static analysis

No suspicious patterns detected.