ffmpeg
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines a bridge to execute
ffmpegandffprobeoperations through a local Python wrapper script located atscripts/media_toolkit.py. This mechanism facilitates the execution of shell commands based on arguments generated by the agent for media processing tasks. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external media files via
ffprobe. Data within these files, such as metadata tags, represents a potential surface for indirect prompt injection where malicious instructions could be embedded to influence the agent's behavior. - Ingestion points: External media files analyzed by the
ffprobebridge (SKILL.md). - Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the workflow.
- Capability inventory: The skill possesses the capability to execute complex
ffmpegandffprobecommands via the Python bridge (SKILL.md). - Sanitization: The skill instructions do not explicitly mention sanitization or validation of the media metadata before it is processed by the agent.
Audit Metadata