video-content-extractor

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a potential for indirect prompt injection as it processes untrusted video content to extract text which is then included in the agent's context. \n
  • Ingestion points: Video files (e.g., lecture.mp4) specified via user parameters. \n
  • Boundary markers: The documentation fails to mention delimiters or specific instructions to the agent to disregard commands found within the OCR results. \n
  • Capability inventory: The skill utilizes ffprobe, ffmpeg, and tesseract for media processing and text extraction. \n
  • Sanitization: No sanitization or validation of the extracted OCR text is described before it is integrated into the final Markdown report. \n- [COMMAND_EXECUTION]: The skill requires the execution of external binaries (ffmpeg, ffprobe, and tesseract) to perform its core functions. \n- [NO_CODE]: The core functionality of the skill is described as residing in a Python script ('scripts/extract_video.py') which is not included in the analyzed files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 09:08 PM
Security Audit — agent-trust-hub — video-content-extractor