The Agent Skills Directory

[COMMAND_EXECUTION]: The recalc.py script uses subprocess.run to invoke the soffice (LibreOffice) binary in headless mode for spreadsheet recalculation.
[REMOTE_CODE_EXECUTION]: The script recalc.py dynamically generates an XML-based StarBasic macro (Module1.xba) and saves it to the user's LibreOffice application configuration directory (~/.config/libreoffice/ or ~/Library/Application Support/LibreOffice/). This macro is subsequently executed to calculate formulas within Excel workbooks.
[COMMAND_EXECUTION]: The documentation in SKILL.md instructs the agent to execute a script named scripts/generate_schematic.py for diagram generation, although this script is not bundled with the provided skill files.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted spreadsheet data.
Ingestion points: External data is loaded using pandas.read_excel() and openpyxl.load_workbook() in both documentation examples and the recalc.py utility.
Boundary markers: There are no markers or specific instructions to isolate the data content from the agent's control logic.
Capability inventory: The agent possesses file system write access and the ability to execute system commands via the provided scripts.
Sanitization: No validation or sanitization is performed on the spreadsheet content before it is processed or used in calculations.

xlsx