ffmpeg
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external media files, creating an attack surface for untrusted data ingestion.
- Ingestion points: Input file paths provided as CLI arguments in gemini.md and gpt.md.
- Boundary markers: Not present (typical for binary media processing).
- Capability inventory: Subprocess calls to FFmpeg and ffprobe in gemini.md and gpt.md.
- Sanitization: Input paths are passed as list elements to subprocess functions, which prevents shell command injection.
- [DYNAMIC_EXECUTION]: The provided Python scripts dynamically generate and execute shell commands to interact with the FFmpeg engine.
- Evidence: Scripts in gemini.md and gpt.md use subprocess.run and asyncio.create_subprocess_exec to perform media operations. This behavior is necessary for the skill's primary function of media automation.
Audit Metadata