ClawHub
Back to skill

Security audit

Voicemaker

Security checks for vulnerabilities and agentic risk

Overview

This Voicemaker skill is mostly coherent, but it needs review because it treats audio generation that may use credits as if it were a safe read action.

Before installing, confirm you are comfortable with agents using your OOMOL-connected Voicemaker account. Treat generate_tts as a user-approved generation or billable action, not as a read-only lookup.

SkillSpector

Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The manifest promises this skill for 'searching and reading data,' but the documented `generate_tts` action creates new content and may consume credits or trigger billable operations. This mismatch can mislead an agent or user into treating the skill as read-only, reducing scrutiny and increasing the chance of unintended non-read execution.

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The safety section says untagged actions are reads, yet `generate_tts` is untagged despite being a generative, stateful operation that can incur usage and produce new artifacts. This unsafe classification may cause autonomous execution without confirmation under a false read-only assumption, leading to unintended spend or action execution.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The instruction to use this skill for 'ANY Voicemaker request' is overly broad and can cause the agent to invoke the skill in cases where direct API use, another tool, or a narrower workflow would be more appropriate. Broad routing language increases the attack surface for accidental invocation and amplifies the effect of the misleading read-only framing elsewhere in the file.

Static analysis

No suspicious patterns detected.