The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of customer reviews and platform profiles. It incorporates specific safety instructions to mitigate potential injection attacks by directing the agent to extract evidence only and ignore any embedded instructions.
Ingestion points: Customer reviews and external profile pages (references/review-programs.md).
Capability inventory: The skill is instructional and does not utilize tools for file system modification, command execution, or network requests.
Sanitization: Explicit defensive instructions are provided to the agent regarding the handling of untrusted third-party content.
[EXTERNAL_DOWNLOADS]: The documentation references external resources hosted on GitHub (github.com/coreyhaines31/marketingskills), including a tool registry and integration guides. These links are provided for informational purposes and do not trigger automatic script execution or package installation.

referrals