frame-extraction

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses FFmpeg via shell commands to perform frame extraction and processing. It also references a 'Local Dependency Bootstrap Rule' from the vendor's shared configuration for environment setup.- [EXTERNAL_DOWNLOADS]: The instructions permit the agent to download video files from TikTok URLs if local copies are unavailable, which involves network access to a well-known external service.- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted data from video manifests and video files.
  • Ingestion points: The skill processes manifest files that map IDs to local video paths and ingests data from local folders (SKILL.md).
  • Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following commands that might be embedded in the metadata of the processed videos or manifests.
  • Capability inventory: The agent possesses capabilities to execute shell commands (ffmpeg) and perform file system read/write operations within the work folder (SKILL.md).
  • Sanitization: The skill does not describe specific sanitization or validation steps for the content of the manifests or video metadata before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 06:37 AM