The Agent Skills Directory

[SAFE]: The skill operates entirely within a 'stage-then-approve' framework. All actions that change the state of the marketer's tool (sending, scheduling, pausing) are explicitly marked as human-only actions. The skill never attempts to execute these actions via its own code or tools.
[SAFE]: Data handled by the skill (email exports, list metrics, credentials) is processed according to best practices. Instructions specifically forbid including sensitive personal data like names or email addresses in run records, focusing only on aggregate counts and segment labels.
[SAFE]: All external downloads and references target well-known and trusted technology services (e.g., Klaviyo, Mailchimp, HubSpot, Google, Airship) or the vendor's own infrastructure (gooseworks-ai). These are documented neutrally as intended parts of the marketing workflow.
[SAFE]: The skill implements robust validation steps, such as the claims-gate and ad-policy-check, ensuring that all marketing copy adheres to brand guidelines and legal requirements before being presented to the user.

email-lifecycle