interpreting-culture-index
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
uv runto execute local Python scripts (scripts/extract_pdf.pyandscripts/check_deps.py) for processing PDF files and verifying the environment. These executions are limited to the skill's own internal logic. - [EXTERNAL_DOWNLOADS]: The skill relies on standard, well-known libraries and system tools (OpenCV, NumPy, Tesseract, Poppler). Instructions for installing these via package managers (brew, apt) are provided for the user.
- [SAFE]: No instances of prompt injection, data exfiltration, or obfuscation were detected. The skill handles PII (names, emails) exclusively for generating assessment reports as per its stated purpose.
- [DATA_EXPOSURE]: While the skill processes personal assessment data, it does not perform any network operations or external data transmissions. All processing is localized to the agent's environment.
Audit Metadata