distill-book
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows established patterns for document processing and agent orchestration. It uses subagents to maintain context hygiene and delegates specialized tasks like PDF processing and skill creation to other tools.
- [INDIRECT_PROMPT_INJECTION]: The skill identifying an attack surface where instructions embedded in a processed book could influence subagent behavior.
- Ingestion points: Document files in various formats (PDF, EPUB, etc.) are read from the environment via file system tools.
- Boundary markers: No explicit use of delimiters or 'ignore embedded instructions' prompts is specified for the subagent interactions.
- Capability inventory: Subagents can write to files; the orchestrator can call the 'skill-creator' skill based on subagent output.
- Sanitization: There is no evidence of filtering or escaping document content before it is processed by the LLM.
Audit Metadata