The Agent Skills Directory

[SAFE]: No security violations, malicious code, or unauthorized data access patterns were identified in the instructions or provided scripts.- [INDIRECT_PROMPT_INJECTION]: The skill features workflows that process external visual data (screenshots and videos) through multimodal vision tools. While this constitutes an indirect prompt injection surface where instructions could be embedded in visual content, the skill includes robust mitigation patterns such as structured evaluation templates, numeric quality scores (≥ 7/10), and manual verification checkpoints.- [DEPENDENCY_ANALYSIS]: The documentation references several internal scripts (gemini_batch_process.py, media_optimizer.py) and an external skill (ui-ux-pro-max) that were not included in the provided file set. These are documented as local dependencies for production-grade generation and asset optimization tasks. The provided scripts (color-contrast-checker.py and responsive-checker.py) are localized utility tools that operate on provided file paths or hex codes without external network access.

frontend-design