ai-voice-cloning

Fail

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the user or agent to install the required command-line interface using curl -fsSL https://cli.inference.sh | sh. This pattern executes a remote script directly in the shell without prior inspection.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx skills add to fetch and install additional modules from the inference-sh/skills repository, introducing external code dependencies into the environment.
  • [COMMAND_EXECUTION]: The skill relies on the infsh CLI tool to execute various audio processing tasks, including voice cloning and media merging, via the Bash tool.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted user data (the text field for voice generation) which is then passed into command-line tools.
  • Ingestion points: User-provided text for TTS generation in SKILL.md examples.
  • Boundary markers: Commands use JSON structures for input which provide some segmentation, but lack explicit instructions to ignore embedded commands.
  • Capability inventory: The skill uses Bash(infsh *) to execute platform-specific commands.
  • Sanitization: No explicit sanitization or escaping of the input text is documented within the skill instructions.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 23, 2026, 10:50 AM