The Agent Skills Directory

[COMMAND_EXECUTION]: The helper script recalc.py programmatically creates a LibreOffice macro file (Module1.xba) in the user's application configuration directory (e.g., ~/.config/libreoffice or ~/Library/Application Support/LibreOffice). This constitutes automated modification of local application settings.
[COMMAND_EXECUTION]: The skill invokes the soffice (LibreOffice) binary via subprocess.run in recalc.py to execute the generated macro for formula recalculation. This involves running external system binaries with parameters derived from agent-managed files.
[PROMPT_INJECTION]: The skill is designed to read and process external spreadsheet data, which introduces a surface for indirect prompt injection.
Ingestion points: Files are read using pd.read_excel() and load_workbook() as described in SKILL.md.
Boundary markers: The instructions lack specific delimiters or instructions to prevent the agent from obeying commands embedded within the data of the spreadsheets being processed.
Capability inventory: The skill possesses capabilities for writing to the filesystem and executing system commands via the provided recalc.py script.
Sanitization: There is no description of data sanitization or validation protocols for the content of the spreadsheet files.

document-skills/xlsx