Security audit
Security checks for vulnerabilities and agentic risk
The skill is mostly a text-to-speech workflow guide, but it asks for broad command and file authority with loose, partly generic scoping.
Review this before installing if your agent grants skills automatic read/write/exec authority. Use it only for explicit TTS or audio-production tasks, keep output paths constrained, avoid passing secrets except the intended license variable, and require confirmation before running any generated shell or Python command.
The description says to use the skill whenever the user wants to improve efficiency, automate workflows, or do batch processing, which is far broader than the stated TTS purpose. Overbroad routing conditions can cause an agent to invoke a powerful skill with read/write/exec tools in unrelated contexts, unnecessarily expanding the attack surface and enabling unintended file or command operations.
The markdown instructs the agent to run scripts and produce outputs, but it does not give prominent user-facing warnings about filesystem writes, external process execution, dependency invocation, or possible overwrites. In practice, this can lead users or orchestrators to trigger actions with system impact without informed consent or awareness of what paths, tools, or data may be affected.
The skill claims that only whitelisted commands should be executed, but elsewhere declares unrestricted exec capability and provides command-oriented usage without defining or enforcing an actual allowlist. In an agent setting, this mismatch can cause operators or downstream models to assume safety controls exist when arbitrary shell execution may still occur, increasing the risk of command injection or unsafe system modification.
