Book Writing
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing user-generated manuscript text, which is a standard ingestion point for indirect instructions. 1. Ingestion points: Manuscript drafts and chapter notes stored in ~/Clawic/data/book-writing/chapters/. 2. Boundary markers: The instructions do not employ specific delimiters to isolate manuscript text from control instructions. 3. Capability inventory: The skill is scoped to read and write files within the local ~/Clawic/data/book-writing/ directory. 4. Sanitization: Manuscript content is processed as provided without specific sanitization filters.
- [SAFE]: The skill follows security best practices by explicitly restricting data storage to a local directory and instructing the agent to avoid saving credentials or personal secrets. All external links point to the author's official website.
Audit Metadata