The Agent Skills Directory

[COMMAND_EXECUTION]: The validation section in SKILL.md uses python3 to execute local linter and validation scripts. These scripts, quick_validate.py and lint_production_skill.py, are located within the $CODEX_HOME environment and are part of the skill's development and verification workflow.
[SAFE]: The skill's primary purpose is to define safety boundaries for AI assistants. It provides structured templates and contracts that enforce best practices, such as source-bounded answers, human-in-the-loop for mutations, and explicit prohibitions against storing secrets or PII.

site-ai-assistant-builder