ClawHub

Mouthpiece

Security checks across malware telemetry and agentic risk

Overview

This is a Chinese workplace advice skill with no code execution or external access; the main caution is that personalization may ask for workplace and financial background.

Install if you want a Chinese-language workplace communication helper. Share only broad context you are comfortable keeping in the agent profile, skip financial-status details unless they are truly useful, and use the update/reset profile instructions if the stored context becomes stale or too personal.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README explicitly asks users to provide personal background such as employer type, job level, and financial status, which can be sensitive profiling data. Because this request appears in onboarding guidance without any privacy warning, minimization guidance, or explanation of storage/retention, users may disclose unnecessary sensitive information that could increase privacy and inference risks.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The activation text is broad enough to match common workplace venting, advice, or reply-writing requests, which can cause the skill to trigger in situations where the user did not explicitly ask for this specialized behavior. Over-broad triggering increases the chance of unintended persona injection, unsolicited guidance, and collection of extra contextual details such as job level or financial status that are not always necessary for the request.

Natural-Language Policy Violations

Medium
Confidence
80% confidence
Finding
The skill is written to operate in Chinese without offering a language choice or documenting that it is limited to Chinese-speaking users. This can cause misunderstanding of advice, misalignment with user expectations, and accidental exclusion or mishandling when invoked for users in other languages.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal