skills/plurigrid/asi/ffmpeg/Gen Agent Trust Hub

ffmpeg

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions designed to override agent behavior or bypass safety guidelines were detected.
  • [Data Exposure & Exfiltration] (SAFE): The skill does not reference sensitive files or include commands to send data to external domains.
  • [Remote Code Execution] (SAFE): No remote scripts are downloaded or executed. The skill only provides local command examples.
  • [Privilege Escalation] (SAFE): No commands for acquiring elevated permissions or modifying system-level settings were found.
  • [Indirect Prompt Injection] (SAFE): While the tool processes external media files and text lists, the skill does not interpolate untrusted data into command structures or prompts in a malicious way.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 08:32 AM