skills/guyronnen/openmontage/ffmpeg/Gen Agent Trust Hub

ffmpeg

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of external media files, which serves as a potential entry point for untrusted data or embedded instructions.
  • Ingestion points: Media files (MP4, GIF, MOV, MP3, etc.) and text-based concat lists (list.txt) processed by ffmpeg and ffprobe commands described in SKILL.md and reference.md.
  • Boundary markers: The instructions do not define delimiters or specific safety instructions to prevent the agent from following directives potentially embedded in media metadata.
  • Capability inventory: The skill utilizes ffmpeg and ffprobe for file system access (reading inputs, writing outputs) and subprocess execution, including the use of the -safe 0 flag which allows processing files from arbitrary paths.
  • Sanitization: There are no recommendations for sanitizing file content or validating filenames before they are passed to shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 12:09 AM
Security Audit — agent-trust-hub — ffmpeg