book-author
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input (genre, core idea, author background) to generate book concepts, outlines, and drafts as defined in Step 1 of
SKILL.md. It lacks explicit boundary markers or instructions to disregard potential commands embedded within this user-provided data. The skill's primary capability includes writing the resulting content to a local markdown file (.md). No specific sanitization or validation routines for the input are described in the instructions.
Audit Metadata