The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing untrusted data from an external system.
Ingestion points: External records are retrieved from ServiceNow tables such as sc_req_item, sc_task, and sc_item_option_mtom using the SN-Query-Table tool within SKILL.md.
Boundary markers: The instructions do not define boundary markers to encapsulate or ignore potential instructions embedded within the retrieved record fields.
Capability inventory: The skill has the capability to modify data in the source system via SN-Update-Record and SN-Add-Work-Notes tools.
Sanitization: There is no explicit sanitization or validation logic for the content fetched from external tables before it is processed in the fulfillment workflow.

request-fulfillment