course-retro
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external feedback data.
- Ingestion points: The skill reads user-provided CSV files for xAPI statements (Step 2) and module feedback (Step 3).
- Boundary markers: None identified. The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the CSV data fields.
- Capability inventory: The skill has the ability to read arbitrary file paths provided by the user and write reports to the
docs/instructional-design/courses/directory. - Sanitization: There is no explicit logic to sanitize, escape, or validate the content of the feedback text before it is processed by the model for clustering or report generation.
- [SAFE]: The skill uses structured local file paths for report generation and does not perform any network operations or remote code execution. External references point to local adapters within the environment root.
Audit Metadata