book-publishing
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external research sources and existing book chapters to calibrate writing styles and content. * Ingestion points: Reads from
content/books/*/chapter-*.mdand external research materials such as Creator Economy reports and historical poetry texts. * Boundary markers: The skill does not define explicit delimiters to separate untrusted external research content from its internal instructions. * Capability inventory: The agent is authorized to write to the filesystem (content/books/), execute project-specific Node.js scripts, and interact with image generation tools. * Sanitization: No explicit content sanitization or validation is described for the ingested research data. - [COMMAND_EXECUTION]: The skill instructs the agent to execute specific shell commands to run local Node.js scripts for document production. * Evidence: Use of
node scripts/generate-book-pdfs.mjsandnode scripts/generate-book-epubs.mjsfor production tasks.
Audit Metadata