The Agent Skills Directory

[PROMPT_INJECTION]: No direct instructions to override safety filters or ignore system prompts were found. The skill processes external data from .claude/product-marketing-context.md and .cursor/product-marketing-context.md, which creates a surface for indirect prompt injection; however, the lack of hazardous capabilities like network access or code execution mitigates this risk. No explicit boundary markers or sanitization steps are defined for these inputs.- [DATA_EXFILTRATION]: The skill identifies and reads specific local project files for marketing context. There are no patterns suggesting the exfiltration of this data to remote servers, nor are there any hardcoded credentials or sensitive system paths accessed.- [SAFE]: The skill consists entirely of markdown-based prompt instructions. It does not include or download any scripts, binaries, or third-party dependencies. References to external tools like brand-visual-generator and logo-generator are internal skill references within the author's ecosystem.

branding