Security audit

音频生成工具-免费版

Security checks across malware telemetry and agentic risk

Overview

This audio skill mostly matches its stated TTS purpose, but it asks agents to run a third-party CLI, persist an API key locally, and uses broader triggers and capabilities than the audio use case needs.

Review this before installing if you may process private scripts, business text, or regulated content. Prefer using an environment variable for DLAZY_API_KEY, avoid sending sensitive text unless you accept dlazy/cloud processing and hosted output URLs, and invoke the skill only for explicit TTS or sound-effect generation rather than general translation, media conversion, or file-processing tasks.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (7)

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The skill presents itself as a narrowly scoped TTS/SFX tool, but later claims broad automation, file processing, API integration, and information retrieval capabilities. This scope expansion increases the chance that an agent will invoke the skill in contexts unrelated to audio generation and grant it more authority than users expect, which can lead to unnecessary command execution or data handling.

Description-Behavior Mismatch

Medium

Confidence: 91% confidence
Finding: The description advertises translation and localization use cases, but the implementation is centered on audio generation through the dlazy CLI. This mismatch can cause an agent or user to route text-processing tasks into a tool that sends content to an external audio service, violating user expectations about function and data handling.

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: Claiming generic information-retrieval capability is unjustified for a TTS/SFX skill and encourages overbroad use. In an agent environment with read/exec/write tools, vague retrieval claims can expand the skill's effective authority and increase the risk of unrelated data access or misuse.

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: The skill claims broad file read/parse/write support beyond what a simple audio-generation wrapper should need. Because the manifest includes read, exec, and write access, overstating file capabilities may cause agents to expose local files or use the skill as a generic file manipulator rather than a constrained TTS tool.

Intent-Code Divergence

Medium

Confidence: 96% confidence
Finding: The skill's guidance is internally inconsistent: it warns against storing API keys in configuration files, yet recommends `dlazy auth set`, which persists the key locally. This can mislead users into placing secrets on disk where they may be exposed through backups, local compromise, or overly permissive file access.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger conditions are overly broad, covering common video, audio, and media-conversion requests rather than narrowly scoping to TTS/SFX generation. This increases the likelihood of accidental invocation in unrelated workflows, exposing user content to external services or unnecessary command execution.

Missing User Warnings

High

Confidence: 98% confidence
Finding: The skill does not clearly disclose up front that user text is sent to a third-party cloud API and that generated audio may be hosted on an external file service. This is a material data-handling omission: users may provide sensitive or proprietary text under the assumption of local processing, creating confidentiality and compliance risk.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.