script
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed of natural language instructions for generating YouTube scripts and metadata. It does not contain any executable code, shell commands, or obfuscated logic.- [DATA_EXPOSURE]: The skill attempts to read context from project-specific files like
.agents/product-marketing-context.md. This is a documented feature designed to help the AI understand the user's business context and does not involve exfiltrating data to external domains.- [INDIRECT_PROMPT_INJECTION]: Although the skill processes data from external files (the product marketing context), it lacks the capabilities (such as network access or shell execution) required to turn an injection into a functional exploit. The output is limited to text generation for the user.
Audit Metadata