local-xlsx-editor
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its ingestion of untrusted spreadsheet data. * Ingestion points: The skill reads user-provided XLSX files from the input/ folder using scripts/local/xlsx_utils.py. * Boundary markers: No delimiters or instructions are present to prevent the agent from interpreting spreadsheet content as instructions. * Capability inventory: The skill executes local Python scripts to read, modify, and write files to the output/ directory. * Sanitization: No data validation or sanitization processes are described for handling cell values or formulas.
- [COMMAND_EXECUTION]: The skill instructs the agent to run specific Python scripts using a local virtual environment (.venv\Scripts\python.exe). This is the primary mechanism for file manipulation and is documented for standard use.
Audit Metadata