Only the latest release gets security fixes.

Anything that makes voice do one of these:

• reach the network outside voice setup;
• read or write files outside the Hugging Face cache, its state folder, its models folder and the clipboard;
• write a transcript anywhere but the clipboard, the terminal voice listen prints to, or the Claude Code prompt voice chat types into;
• type into Claude Code anything other than what you said;
• run code from a model, or load a model family other than Kokoro and Parakeet;
• signal a process that is not its own.

A known vulnerability in a locked dependency counts too.

Report it privately on GitHub: on the repo's Security tab, choose Report a vulnerability. Include what you found, how to reproduce it, and what an attacker could do with it.

The maintainer replies in the private advisory, works out the fix there, and publishes the advisory with the release that fixes it.