kai-motion-spot
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill performs extensive local command execution using Python scripts located in the AI agent's installation directory (e.g.,
render.py,assets.py,music.py,upload_post.py). These scripts handle tasks ranging from video rendering to asset manipulation and automated social media posting. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data such as product descriptions from
MARKETING.md, customer quotes, and user recordings. This content is interpolated into generated HTML and video captions, creating an attack surface for indirect prompt injection. - [DYNAMIC_EXECUTION]: The skill dynamically generates HTML/JavaScript (
index.html) and structured JSON specifications. These files are then executed or parsed by rendering engines (Playwright) and local scripts to produce the final video product. - [CREDENTIALS_UNSAFE]: The skill references sensitive API keys (
ELEVENLABS_API_KEY,UPLOAD_POST_API_KEY). It follows security best practices by explicitly instructing the agent to fetch these from the environment only and prohibiting them from being printed, stored in files, or pasted into chat.
Audit Metadata