The Agent Skills Directory

[COMMAND_EXECUTION]: The recalc.py script executes the soffice binary to perform formula recalculation. The script uses a list of arguments in subprocess.run instead of a shell string, which is a safer practice that mitigates risks of command injection.
[DYNAMIC_EXECUTION]: The skill dynamically generates a LibreOffice Basic macro file (Module1.xba) in the user's local configuration directory. This macro is hardcoded and restricted to workbook calculation and saving tasks necessary for the skill's operation.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process spreadsheet files, which creates an attack surface for indirect prompt injection.
Ingestion points: External spreadsheets are ingested using pandas.read_excel and openpyxl.load_workbook as specified in the skill documentation.
Boundary markers: The skill does not provide specific instructions or delimiters to the agent to distinguish spreadsheet data from potential instructions.
Capability inventory: The skill has the ability to read/write files and execute shell commands to manage spreadsheet recalculations.
Sanitization: No explicit sanitization or validation is performed on the content read from spreadsheet cells.

xlsx