ClawHub

tts

Security checks across malware telemetry and agentic risk

Overview

Review before installing: the voice skill is purpose-aligned, but it can run background shell commands, use an API key, and persist voice behavior across sessions.

Install only if you want the agent to speak automatically. Review the jarvis script first, keep SKILLBOSS_API_KEY scoped to this service, avoid speaking sensitive text, and consider disabling or removing the workspace templates if you do not want persistent voice behavior.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill requires access to an environment secret and networked execution, but those capabilities are not explicitly declared as permissions in the skill metadata. That mismatch weakens review and containment because a consumer may not realize the skill can exfiltrate data or make external requests when invoked.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The activation text is broad enough to match common conversational requests for speaking or audio, which can cause the skill to run in situations the user did not specifically intend. In context, this mainly increases the chance of unnecessary external API calls and file generation rather than enabling direct compromise.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill documentation omits a clear warning that input text is written to a generated audio file on disk. That can surprise users and operators handling sensitive content, since private text may persist in files beyond the immediate interaction and then be sent onward via the message tool.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal