The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown and reference guides for digital marketing workflows. It contains no executable code, shell scripts, or automated network operations.
[DATA_EXPOSURE_AND_EXFILTRATION]: No evidence of hardcoded credentials or unauthorized data harvesting. The guide explicitly warns users that hashing emails does not anonymize personal data and emphasizes the need for approved data rights and consent when uploading audience lists.
[INDIRECT_PROMPT_INJECTION]: While the skill involves processing external data (such as landing page URLs and product descriptions), it is used to generate creative copy and marketing strategies under user supervision. The instructions mandate verifying authorization and account access before taking any action, providing a human-in-the-loop safety boundary.
[PROMPT_INJECTION]: No instructions were found that attempt to bypass safety filters or override agent behavior. Hypothetical benchmarks and testimonials are explicitly labeled as synthetic to prevent misleading the agent or user.

paid-ads