ClawHub

speack2u

Security checks across malware telemetry and agentic risk

Overview

This TTS personalization skill is coherent and instruction-only, but users should understand it may save voice preferences and references optional paid API keys.

Install only if you are comfortable with the agent remembering voice and speaking-style preferences. Review any saved Voice, Style, Spoken Text, and Avoid entries, confirm TTS configuration changes yourself, and do not paste real API keys into shared files, logs, or screenshots.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
Findings (4)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description says it 'auto-evolves' and learns from user feedback, but it does not clearly define when or how that behavior is activated. Ambiguous activation around adaptive behavior can cause the agent to collect preferences or modify TTS settings without clear user awareness, increasing the risk of unexpected personalization and silent state changes.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly instructs itself to detect patterns from user feedback and adapt TTS configuration, but it provides no warning to the user that their feedback may be observed and used to update persistent settings. In context, this makes the skill more dangerous because it is designed for ongoing behavioral adaptation, which can create privacy, consent, and transparency issues even if the adaptation is limited to TTS preferences.

Missing User Warnings

Low
Confidence
86% confidence
Finding
The configuration examples include live secret fields such as `openai.apiKey` and `elevenlabs.apiKey` without any guidance on secure secret handling. This can lead users to paste credentials directly into shared config files, documentation snippets, logs, or version control, increasing the risk of accidental key disclosure and subsequent unauthorized API usage.

Self-Modification

High
Category
Rogue Agent
Content
# Criteria for Voice Preferences / 语音偏好判断标准

Reference only — consult when deciding whether to update SKILL.md.
仅供参考 — 决定是否更新 SKILL.md 时查阅。

## When to Add / 何时添加
Confidence
89% confidence
Finding
update SKILL

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal