video-scene-analyzer
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes ffmpeg, ffprobe, and a bundled shell script (scripts/videx) to extract and analyze video frames.
- [DYNAMIC_EXECUTION]: Executes a local bash script included in the skill package to manage the video processing workflow.
- [INDIRECT_PROMPT_INJECTION]: 1. Ingestion points: Ingests visual content and metadata from untrusted video files provided by the user (processed via ffmpeg and ffprobe). 2. Boundary markers: The instructions include a structured JSON schema for output and specific transcription guidelines, which provide some implicit framing but no formal sandboxing of ingested text. 3. Capability inventory: The skill has the ability to execute shell commands (ffmpeg, ffprobe, videx), read local image files, and write to the local file system. 4. Sanitization: There is no explicit sanitization or filtering of the text transcribed from the video frames before it is processed by the agent.
Audit Metadata