The Agent Skills Directory

[COMMAND_EXECUTION]: The recalc.py script executes the local soffice (LibreOffice) binary to recalculate formulas. It programmatically creates a LibreOffice macro file in the user's application configuration directory to facilitate this process. This behavior is documented and consistent with the skill's stated purpose of handling complex spreadsheet formulas.
[PROMPT_INJECTION]: The skill processes untrusted data from external spreadsheets, creating a potential vector for indirect prompt injection if malicious instructions are embedded within cells.
Ingestion points: The skill uses pandas.read_excel and openpyxl.load_workbook to read user-supplied Excel, CSV, and TSV files.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are implemented to separate cell data from the agent's internal reasoning.
Capability inventory: The agent has permissions to write files to the filesystem and execute the soffice binary through the provided recalc.py script.
Sanitization: The skill lacks explicit sanitization or validation of spreadsheet content before it is incorporated into the agent's context.

xlsx