The Agent Skills Directory

[SAFE]: The skill consists entirely of informational Markdown documents and data templates (JSON, CSV). No executable scripts, binaries, or automated tool configurations are present in the skill package.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external inputs such as brand rules and product-level visual requirements. While this technically creates a surface for indirect prompt injection, the skill possesses no capabilities (such as shell access, file-writing, or network operations) that would enable a malicious payload to compromise the environment.

design-tokens