video-explorer
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill documentation instructs the agent to execute a local shell script (
videx) with variables provided by the user or extracted from content (e.g.,<video>,<start>-<end>). This pattern creates a potential command injection surface if the inputs are not strictly sanitized by the agent or the script.- [PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection (Category 8) because it processes untrusted external data (video files). - Ingestion points: Video files provided for analysis.
- Boundary markers: Absent; there are no instructions for the agent to ignore or distinguish between the video content and its own instructions.
- Capability inventory: Local shell execution via
./skills/video-explorer/scripts/videxand file system writes to./videx-out/. - Sanitization: None specified in the markdown; the internal sanitization of the script is unverifiable.
Audit Metadata