audio-generation

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands and CLI tools based on user input and specific workflow states.
  • Evidence: "Choose the smallest matching command or workflow from the user input and run it directly."
  • Evidence: "run postplus quote confirm --json --challenge-file <challenge.json> and retry with the returned token."
  • [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted user data (TTS scripts, task descriptions, and persona requirements) and uses them to determine which commands to execute, creating a vulnerability surface.
  • Ingestion points: User requests for text-to-speech (TTS), voice cloning, and translation dubbing defined in SKILL.md.
  • Boundary markers: The skill defines a "Core Boundary" and "Public Command Boundary" to restrict its actions to classification and handoff.
  • Capability inventory: The skill triggers CLI commands (postplus) and scripts as part of its routing logic.
  • Sanitization: The instructions do not specify sanitization or validation of the user-provided content before it influences command selection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:59 AM
Security Audit — agent-trust-hub — audio-generation