Skills
skills/veniceai/skills/venice-video/Gen Agent Trust Hub

venice-video

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of external content through YouTube transcription and media URL processing, which creates a surface for indirect prompt injection.\n
  • Ingestion points: YouTube audio via the /video/transcriptions endpoint and external media files via image_url, video_url, and reference_image_urls as described in SKILL.md.\n
  • Boundary markers: Absent; the documentation does not specify the use of delimiters or instructions to prevent the agent from obeying commands embedded in transcribed text.\n
  • Capability inventory: The skill allows the agent to retrieve and process external content which is then returned to the active context, potentially influencing downstream logic.\n
  • Sanitization: No sanitization or filtering of transcribed text is specified.\n- [COMMAND_EXECUTION]: The skill documentation includes standard curl command examples for interacting with the Venice AI API endpoints (e.g., api.venice.ai/api/v1/video/queue). These are provided as implementation references and target the vendor's official infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 08:03 PM