The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from external spreadsheet files (.xlsx, .xlsm, .csv, .tsv), which presents an indirect prompt injection surface.
Ingestion points: Spreadsheet data enters the agent context through pd.read_excel() and load_workbook() operations in the programmatic approach, as well as through browser-based reading of Excel Online workbooks.
Boundary markers: The instructions lack explicit boundary markers or directives to ignore instructions that might be embedded within the spreadsheet data.
Capability inventory: The skill has the capability to write files to the disk, execute a local Python script via subprocess (scripts/recalc.py), and perform browser automation through Playwright (including actions like creating charts and pivot tables).
Sanitization: There is no mention of sanitizing, escaping, or validating the content of external spreadsheet cells before processing the data or using it in decision-making logic.
[COMMAND_EXECUTION]: The skill instructs the agent to execute a local script scripts/recalc.py using a subprocess call to perform formula recalculation via LibreOffice.

xlsx