ffmpeg-cli

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a vast collection of FFmpeg shell commands as examples for media conversion, audio mixing, and video filtering. These are instructional templates for the agent to follow when performing media-related tasks.
  • [PRIVILEGE_ESCALATION]: Installation instructions in SKILL.md include the use of sudo apt install ffmpeg. This is standard practice for installing software via system package managers and does not indicate malicious intent.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes functionality for reading content from external text and subtitle files using the drawtext and subtitles filters, which represents a potential attack surface for untrusted data ingestion.
  • Ingestion points: External files such as text.txt and subs.srt used in FFmpeg filters in references/advanced-editing.md.
  • Boundary markers: No delimiters or safety instructions are provided to the agent to ignore potentially malicious content within these files.
  • Capability inventory: The skill enables tools such as Bash, Read, Glob, Grep, Write, and Edit for interacting with the environment.
  • Sanitization: No sanitization or validation steps are mentioned for the content of processed text or subtitle files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 07:00 AM