ClawHub
Back to skill

Security audit

Dlazy Gemini 2.5 Tts

Security checks across malware telemetry and agentic risk

Overview

The skill is a disclosed cloud text-to-speech wrapper using dLazy's CLI, with a documentation output-format mistake but no hidden or destructive behavior found.

Install only if you are comfortable sending prompts to dLazy's hosted service and storing a dLazy API key locally. Use the npx alternative if you do not want a global CLI, avoid passing private local files unless you intend to upload them, and do not rely on the shown output schema until the publisher fixes the audio/image mismatch.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (1)

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The documented output format conflicts with the skill’s stated text-to-speech purpose by showing an image result (`type: image`, PNG MIME type, `result.png`). This can mislead an agent or user into handling outputs incorrectly, causing unsafe automation decisions, broken downstream parsing, or accidental acceptance of the wrong artifact type from a remote service.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.