xlsx
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The
recalc.pyscript dynamically generates a StarBasic macro and writes it to the user's local LibreOffice configuration directory (e.g.,~/.config/libreoffice/...). This macro is then executed via thesofficecommand-line interface to perform formula recalculations that are not supported by the Pythonopenpyxllibrary alone. - [COMMAND_EXECUTION]: The
recalc.pyscript usessubprocess.runto execute thesoffice(LibreOffice) binary and system utilities liketimeoutorgtimeout. These commands are used to automate the opening, recalculating, and saving of Excel workbooks. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process data from external Excel files, which represents a surface for indirect prompt injection. Malicious data or formulas within a spreadsheet could potentially influence the agent's reasoning or actions.
- Ingestion points: Data is loaded from
.xlsx,.xlsm,.csv, and.tsvfiles usingpandasandopenpyxlas described inSKILL.md. - Boundary markers: No specific delimiters or "ignore" instructions are provided to the agent for handling content read from spreadsheet cells.
- Capability inventory: The skill possesses file system access, shell command execution (via
recalc.py), and the ability to generate documentation containing URLs. - Sanitization: There is no evidence of sanitization or validation logic for external data processed by the skill.
Audit Metadata