The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external spreadsheet files (.xlsx, .xlsm, .csv, .tsv), which introduces a surface for indirect prompt injection if the files contain malicious instructions.
Ingestion points: Untrusted data enters the agent context through file-read operations described in the Reading and analyzing data section using pd.read_excel() and load_workbook().
Boundary markers: No specific boundary markers or instructions to ignore embedded content are provided for the data ingestion process.
Capability inventory: The skill possesses file-writing capabilities (wb.save, df.to_excel) and the ability to execute shell commands (python scripts/recalc.py).
Sanitization: There is no evidence of data sanitization or validation routines for the content extracted from processed spreadsheets.
[COMMAND_EXECUTION]: The skill instructs the agent to perform shell command execution to recalculate Excel formulas using a bundled script.
Evidence: The Recalculating formulas section explicitly directs the execution of python scripts/recalc.py <excel_file>. The script is also noted to perform configuration tasks for LibreOffice on its first run.

xlsx