The Agent Skills Directory

Policy Extraction from Conversation History: The skill analyzes conversation history to generate automation hooks in the .github/hooks/ directory. While this enables context-aware automation, it relies on conversational data entering the agent context (from SKILL.md) without explicit boundary markers or sanitization steps. Users should be aware that conversational inputs directly influence the logic of the generated scripts or policy files.
Automated Script and Hook Generation: This tool is designed to draft executable scripts and JSON configuration files to automate agent events. The skill possesses the capability to write these files based on the extraction of user requirements. Because the generated code is derived from conversational context, it is a best practice to review the drafted logic to ensure it accurately implements the intended policy without introducing unintended automation behaviors.

create-hook