The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from ServiceNow records and interpolates it into email templates, creating a surface for indirect injection if the source data contains malicious instructions.
Ingestion points: Data is ingested from sysevent, incident, and sys_journal_field tables via the current object and GlideRecord queries.
Boundary markers: The provided examples do not use explicit delimiters or safety instructions to isolate untrusted data from the notification logic.
Capability inventory: The skill utilizes snow_execute_script for server-side logic and tools to manage email configurations and artifacts.
Sanitization: The example scripts do not demonstrate HTML escaping or input validation before rendering data into the email body.
[DYNAMIC_EXECUTION]: The skill facilitates the creation and management of executable scripts within the ServiceNow environment, such as Mail Scripts (sys_script_email) and Business Rules.
Evidence: The documentation provides multiple JavaScript (ES5) templates designed to be stored and executed by the ServiceNow platform for dynamic content generation and recipient routing.

email-notifications