The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes Python scripts (scripts/export_expert_report.py and scripts/validate_artifacts.py) that utilize the subprocess module to execute local shell commands. These commands invoke standard document utility tools such as pandoc for format conversion and system browsers (Chrome, Edge, or Chromium) in headless mode for PDF generation. The implementation follows security best practices by passing command arguments as a list rather than using shell execution, which mitigates the risk of command injection.
[SAFE]: The skill demonstrates a high level of security awareness through the inclusion of scripts/validate_artifacts.py. This script performs automated hygiene checks on generated reports, specifically scanning for and flagging the presence of local filesystem paths (e.g., /Users/ or file://) and unresolved placeholders. This proactive measure ensures that sensitive environment metadata from the host machine is not leaked into the final public-facing deliverables.
[SAFE]: Analysis of the skill's instructions in SKILL.md and references/ shows a consistent focus on educational and research goals. The workflow is transparent, and the tool routing guidelines explicitly exclude high-stakes advice (legal, medical, or safety), emphasizing its use for educational research with explicit source and uncertainty labeling.

yao-expert-skill