The Agent Skills Directory

[NO_CODE]: The skill contains no executable code, scripts, or command-line instructions. It functions exclusively as a set of natural language guidelines for the AI agent.
[SAFE]: No patterns of prompt injection, data exfiltration, or obfuscation were found. The workflow focuses on UI/UX transformation tasks like mapping components and updating CSS tokens.
[SAFE]: While the skill instructions involve processing external data (Current HTML/CSS/JS), the skill itself does not define any capabilities to execute code, write to the filesystem, or communicate over the network, effectively neutralizing the surface for indirect prompt injection.

krds-transform