tts
Warn
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation directs the agent to download source code and install dependencies from a third-party GitHub repository (
github.com/3Craft/tts) which is not a recognized trusted organization. - [REMOTE_CODE_EXECUTION]: The setup instructions involve cloning an external repository and performing a
pipx install, resulting in the execution of third-party code that is not verified by the platform. - [COMMAND_EXECUTION]: The wrapper script
scripts/vox_tts.pyutilizessubprocess.runto invoke the externalvoxCLI. While it correctly uses argument lists to mitigate common shell injection vulnerabilities, it facilitates the execution of external binaries with user-influenced parameters. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted text and audio data, creating a potential attack surface where malicious input could influence the behavior of the underlying processing models or the CLI tool.
- Ingestion points: User-supplied text for synthesis via the
speakanddesigncommands, and external audio/text files for thetranscribeandbatchcommands inscripts/vox_tts.py. - Boundary markers: The skill instructions do not specify any delimiters or safety warnings to ensure the agent ignores instructions embedded within the data being processed.
- Capability inventory: The skill is capable of executing shell commands and performing file system write operations.
- Sanitization: There is no evidence of input validation or content filtering before data is passed to the external processing utility.
Audit Metadata