The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data from review platforms (Google, Trustpilot, G2, Capterra) which presents an attack surface for indirect prompt injection. The ingestion points include fetched customer reviews and profiles. The skill includes a mitigation instruction in references/review-programs.md advising the agent to ignore instructions embedded in fetched text. However, no technical sanitization is described for these inputs. The skill's capabilities include drafting public responses and interacting with referral tools.
[EXTERNAL_DOWNLOADS]: The skill references implementation guides and a tool registry hosted on a public GitHub repository at github.com/coreyhaines31/marketingskills. These references provide documentation for tool integrations and target a well-known service.

referrals