xlsx
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Indirect Prompt Injection] (SAFE): The skill's primary function is to process external data from CSV and Excel files, which inherently creates a surface for indirect prompt injection. This is evaluated as safe because it is necessary for the skill's primary purpose and no exploitation logic is present. Evidence Chain: Ingestion points are located in examples/data-transformation.md and references/best-practices.md; boundary markers are absent in the examples; capabilities include filesystem write access via openpyxl; sanitization logic is not provided in the examples.
- [Unverifiable Dependencies & Remote Code Execution] (SAFE): All referenced dependencies (openpyxl, pandas, xlsxwriter, xlrd) are industry-standard, trusted packages. Installation is handled via standard package managers. No remote code execution patterns or suspicious download behaviors were detected.
- [Metadata Poisoning] (SAFE): Metadata and documentation are clear and correspond to the skill's intended behavior. A hardcoded local path in the README is a minor best-practice violation but poses no security risk. Note: The implementation script 'scripts/excel_helper.py' referenced in the documentation is missing from the provided file bundle.
Audit Metadata