The Agent Skills Directory

[SAFE]: The skill references established open-source repositories DietrichGebert/ponytail and microsoft/SkillOpt for best practices in goal optimization and implementation review. These are well-known references in the AI agent community.
[COMMAND_EXECUTION]: The skill includes local Python scripts (scripts/goal_state.py and scripts/validate_goal_file.py) to manage task states and validate file formats. These scripts utilize standard libraries for file parsing and do not exhibit malicious behavior such as network exfiltration or credential harvesting.
[SAFE]: The skill implements a robust safety framework by requiring explicit user confirmation before executing goals or performing destructive operations. It uses a 'draft' status by default to prevent unauthorized execution.

auto-goal