The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional text designed to guide an AI agent in matching existing codebase conventions. It does not include any scripts, executable commands, or configuration that would allow for unauthorized access or code execution.
[SAFE]: The workflow explicitly promotes secure coding practices by instructing the agent to 'Never imitate a precedent that is unsafe, incorrect, or incompatible with the requested behavior' (Workflow Step 6).
[SAFE]: The instructions require the agent to seek explicit user authorization before introducing novel constructs or new dependencies, ensuring human-in-the-loop oversight for significant changes.

plexy-sheep