xlsx
Pass
Audited by Gen Agent Trust Hub on May 21, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted spreadsheet data (.xlsx, .csv, .tsv) through
pandasandopenpyxl, creating an indirect prompt injection surface. A malicious file could contain instructions designed to manipulate the agent's behavior during data analysis or formatting tasks.\n - Ingestion points:
pd.read_excel()andload_workbook()calls inSKILL.mdindicate processing of external files.\n - Boundary markers: No specific delimiters or instructions are provided to the agent to treat input data as non-instructional.\n
- Capability inventory: The skill possesses the ability to write to the file system (
wb.save) and execute a local Python script (scripts/recalc.py).\n - Sanitization: No input validation or content sanitization is described for the spreadsheet data.\n- [COMMAND_EXECUTION]: The skill requires the execution of a local utility script,
scripts/recalc.py, to perform formula recalculations and manage background LibreOffice configurations.\n - Evidence: The workflow instruction
python scripts/recalc.py output.xlsxinSKILL.mddirects the agent to run a shell command.
Audit Metadata