The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION] (LOW): The skill is designed to ingest and process external datasets. While this presents a potential injection surface, the skill's capabilities are restricted to read-only validation tasks (using Read, Grep, and Glob) and it implements strict schema validation and length constraints that serve as effective sanitization barriers.\n- [COMMAND_EXECUTION] (LOW): The skill references local Python scripts and provides shell script templates for pre-commit hooks (e.g., poetry run). These are standard development practices for maintaining data quality and do not constitute a security risk in a developer-centric environment.\n- [DATA_EXFILTRATION] (SAFE): No network operations or external data transmissions were found. The skill includes a 'forbidden URL' list that explicitly flags localhost and 127.0.0.1, which is a proactive security measure to prevent accidental exposure of local services during dataset validation.\n- [REMOTE_CODE_EXECUTION] (SAFE): All logic is local. There are no patterns involving the download or execution of remote scripts or unverified third-party code.

golden-dataset-validation