course-content-authoring
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied outlines and material definitions to produce output files.- Ingestion points: Content is derived from day{n}/outline.md and interactive user input during the content authoring process.- Boundary markers: Absent; no specific delimiters or instructions to treat user data as untrusted are used when handling input.- Capability inventory: The skill performs file system write operations, creating .md, .csv, and .yaml files locally.- Sanitization: Absent; there is no mention of filtering or validating input strings before they are written to course artifacts.- [SAFE]: The skill provides helpful implementation details, such as the correct application of a UTF-8 BOM for CSV files to ensure compatibility with spreadsheet software, and clear instructions for maintaining data integrity across different stages of course development.
Audit Metadata