The Agent Skills Directory

[SAFE]: The skill provides a robust set of templates and local scripts for brand documentation. The Python scripts (brand-book and generate) use standard libraries for file operations and data parsing.
[INDIRECT_PROMPT_INJECTION]: The skill processes brand data from markdown templates to generate image prompts. The ingestion points are the markdown files, and capabilities are limited to prompt construction for stdout. Boundary markers are present in the output, and sanitization is performed through regex frontmatter extraction. This attack surface is considered safe given the skill's primary purpose and limited execution scope.

brand-designer