The Agent Skills Directory

[SAFE]: The skill implements standard best practices for email authentication and domain warmup, focusing on legitimate outbound operations and reputation management.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data regarding sending domains, volume metrics, and deliverability history.
Ingestion points: User inputs collected during the intake phase in SKILL.md.
Boundary markers: The skill does not define specific isolation markers for user data.
Capability inventory: Includes local file reading via a validation script and instructions for DNS diagnostics.
Sanitization: User-provided strings are used for configuration purposes without explicit sanitization steps.
[COMMAND_EXECUTION]: The skill suggests using standard network utilities like dig for DNS record verification. These are used for auditing the user's own infrastructure and are not executed in a way that allows arbitrary command injection.
[EXTERNAL_DOWNLOADS]: The instructions reference reputable third-party services such as Google Postmaster Tools, Microsoft SNDS, and MXToolbox. These are well-known services for email reputation auditing and are considered safe sources.

email-deliverability