The Agent Skills Directory

[SAFE]: A comprehensive review of the skill's markdown files and evaluation logic reveals no evidence of prompt injection, obfuscation, or malicious command execution. The content is strictly informational and advisory.
[DATA_EXPOSURE]: The skill instructions include reading from local context files such as .agents/product-marketing-context.md. These are project-specific configuration files used to maintain state and do not represent exposure of sensitive system credentials or private user data.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes data from external context files and user input. However, because the skill does not utilize high-risk tools (e.g., shell access, file writing, or network requests), the risk associated with this ingestion point is negligible.

referral-program