The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection via external API responses and webhooks. It proactively mitigates this risk by providing rigorous instructions for input validation, data sanitization, and signature verification before any external data enters the system's business logic (references/10-security.md).
[DATA_EXFILTRATION]: The skill architecture centralizes outbound communication through an Anti-Corruption Layer (ACL). It emphasizes security best practices by mandating the use of secrets managers for credentials and prohibiting the logging of sensitive information like API keys or personal data (references/03-api-client.md, references/07-authentication.md).

integrations