chinese-voice-skill

• Outcome: Pass
• Latest audit: May 29, 2026, 8:20 PM
• Version:
  1.0.4

Overview

This appears to be a voice-reply/TTS skill whose main risk is overly broad auto-activation, not evidence of malicious behavior.

Install this only if you want the agent to enter voice/TTS mode from voice-related prompts. Prefer explicit opt-in wording and confirm before sending sensitive text to any external TTS service.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The skill defines many broad read_when phrases such as '开始语音', '发送语音', and '语音对话', which can overlap with ordinary conversation or unrelated requests. Because the skill is configured to auto-trigger, ambiguous matching can cause unintended execution, unnecessary tool invocation, and accidental transmission of user content to an external TTS service.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The description states the skill 'auto-triggers when the user needs a voice reply' but does not define precise boundaries for what constitutes that need. In an agent system, vague auto-activation criteria increase the chance of unexpected skill invocation and processing of content the user did not intend to convert to speech.

VirusTotal

65/65 vendors flagged this skill as clean.