The Agent Skills Directory

[SAFE]: The skill implements a legitimate productivity tool for financial organization. It uses standard Python modules like csv, json, and argparse to process local files without any suspicious behavior.
[DATA_EXPOSURE]: Although the skill processes sensitive financial data, all operations are local. The script scripts/invoice_categorizer.py reads user-provided CSV files but does not include any network functions or exfiltration logic. No hardcoded credentials or sensitive system paths were detected.
[COMMAND_EXECUTION]: No instances of arbitrary command execution, shell injection vulnerabilities, or unsafe execution of dynamic code (such as eval() or exec()) were found in the provided scripts.
[REMOTE_CODE_EXECUTION]: The skill does not define any external Python or Node.js dependencies and contains no logic to download or execute scripts from remote servers.
[PROMPT_INJECTION]: The instructions in SKILL.md are focused on workflow guidance and do not contain patterns designed to override agent safety protocols or system instructions.

invoice-organizer