book-chapter-generator
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to create a directory hierarchy under
/docs/chapters/and write individualindex.mdfiles for each chapter using shell commands. This is standard behavior for structural generation skills.\n- [DYNAMIC_EXECUTION]: The skill provides Python code snippets for the agent to perform graph analysis, including topological sorting for dependency validation and extracting Concept Impact Scores (CIS). These scripts are intended for internal validation but represent a surface for dynamic code logic execution.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local files (course-description.md,learning-graph.json) to generate user-facing chapter content.\n - Ingestion points: Reads course goals and concept labels from local documentation and JSON graph data.\n
- Boundary markers: None explicitly provided to separate untrusted data from the agent's generation instructions.\n
- Capability inventory: File creation, directory creation (
mkdir), and modification of themkdocs.ymlconfiguration file.\n - Sanitization: The skill enforces strict formatting for URL path names (lowercase and dashes only), which mitigates some injection risks during file creation.
Audit Metadata