kie-external-api

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is assessed as safe, with all activities directly supporting its stated purpose of AI media generation and analysis.
  • [COMMAND_EXECUTION]: The skill uses local commands like ffmpeg and ffprobe to extract frames and audio from user-provided video files. These operations are essential for the style analysis and cloning features and do not involve unauthorized system access.
  • [EXTERNAL_DOWNLOADS]: The skill recommends the use of trusted third-party software, including ffmpeg and OpenAI's whisper model. These are standard tools in the AI and media industry and are sourced from official package registries.
  • [PROMPT_INJECTION]: The skill ingests user media for processing and transcribes audio to assist in prompt creation. While this introduces a surface for indirect prompt injection, the risk is managed within the context of generating creative assets through the KIE.ai API.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 07:52 AM