synthesize-sfx-ffmpeg

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on executing ffmpeg and ffprobe via the shell to generate sound effects and verify their properties. It constructs complex filter graphs at runtime to define the audio characteristics.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting user-defined parameters which are then used to build shell commands.
  • Ingestion points: Untrusted data enters the context through the SFX type, Duration, Output directory, and Custom specs (frequency, harmonics, decay rates, noise color) provided in the user brief.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat these inputs as data only or to ignore embedded instructions within them.
  • Capability inventory: The skill utilizes shell execution for ffmpeg and ffprobe as seen in the workflow and the verification scripts (tests/verifier.md).
  • Sanitization: No sanitization, escaping, or validation logic is specified to prevent malicious inputs from breaking out of the ffmpeg filter string or the shell command context.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:06 PM
Security Audit — agent-trust-hub — synthesize-sfx-ffmpeg