The Agent Skills Directory

[SAFE]: The skill implements a notification framework using standard architectural patterns. Access control is managed through Row-Level Security (RLS) on public tables such as notifications and notification_preferences, while sensitive operational data like delivery_log and notification_suppressions (containing PII like email addresses and phone numbers) is correctly isolated in private schemas not exposed to the API layer.- [EXTERNAL_DOWNLOADS]: The skill utilizes the @constructive-io/node-type-registry Node.js package. This is an official resource provided by the author for module configuration and does not represent an untrusted dependency.- [INDIRECT_PROMPT_INJECTION]: The notification system ingests data from external sources including provider webhooks (Amazon SES, SendGrid, Twilio) and notification content stored in the notifications table. While this ingestion represents a surface for indirect prompt injection, it is the primary intended function of the skill, and the documentation emphasizes proper correlation and audit logging.

constructive-notifications