review-cet-writing
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts (
scripts/convert_score.pyandscripts/generate_report.py) to process score data and render HTML reports. These scripts use standard Python libraries, operate on local files, and do not perform any remote execution or unauthorized command injection. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes user-provided text (essays and translations) to generate feedback, creating a standard surface for indirect prompt injection.
- Ingestion points: User-provided writing or translation text processed via the workflow in
SKILL.md. - Boundary markers: The instructions do not define specific delimiters to isolate user text from system grading instructions.
- Capability inventory: The skill can execute local scripts and write files to generate HTML reports.
- Sanitization: The report generation script (
scripts/generate_report.py) directly interpolates text highlights into the HTML template without explicit escaping, which is a potential point for content injection from the user text to affect the final report rendering.
Audit Metadata