ClawHub
Back to skill

Security audit

beware-piper-tts-free

Security checks for vulnerabilities and agentic risk

Overview

This local TTS skill has plausible core behavior, but its instructions mix local-only claims with API-key, callback, and external-service language that users should review before installing.

Review this skill carefully before installing. Do not provide API keys or callback URLs unless the publisher clarifies what service is used and what data is sent. Run it only in an environment where local command execution, package installation, model downloads, and MP3 file writes are acceptable.

SkillSpector

Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The documentation makes mutually inconsistent security claims: it says the skill is fully local with no cloud calls and no API key, but also instructs users to configure an API key and later references API/auth and HTTP failure modes. This is dangerous because users and agents may make trust decisions based on the 'local/no-key' claim while the skill behavior or its associated scripts may actually invoke remote services or require secrets, increasing the risk of secret exposure and unintended data egress.

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The file markets the free version as limited to a single default voice and no style control or long-text segmentation, but later functional sections describe options such as pitch/speed controls, multi-voice-related handling, and long-text support patterns. These contradictions can mislead an agent into attempting unsupported execution paths or trusting generic boilerplate that may not match the real implementation, which is a supply-chain/documentation integrity issue.

Context-Inappropriate Capability

Medium
Confidence
83% confidence
Finding
The skill is presented as a local TTS utility, yet it advertises generic API integration and external service interaction capabilities unrelated to that purpose. Unnecessary claimed capabilities broaden the apparent trust boundary and may normalize network access or external data transfer in a skill that users expect to remain local-only.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill describes installation, command execution, and writing generated MP3 files to local paths, but does not provide a clear upfront warning about execution side effects, filesystem writes, downloads, or dependency installation. This is dangerous because users may invoke the skill expecting a simple text transformation while it performs system changes or writes files, increasing the risk of unintended modification of the environment.

Static analysis

No suspicious patterns detected.