problem-management
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from ServiceNow incident records which acts as an entry point for untrusted content.
- Ingestion points: Fields from the
incidenttable (short_description,category,subcategory,cmdb_ci) are read inKEDBSearch.findForIncidentandcreateProblemFromIncidentsinSKILL.md. - Boundary markers: None are implemented in the JavaScript snippets to separate record data from execution logic.
- Capability inventory: The skill uses the
snow_execute_scripttool to run code capable of creating and updating database records (problem.insert(),incident.update(),known_error.insert()). - Sanitization: The code performs basic string splitting for keyword searches but does not include formal sanitization of incident content before it is stored in new records or used in queries.
- [DYNAMIC_EXECUTION]: The skill utilizes the
snow_execute_scripttool to execute server-side JavaScript (ES5) within the ServiceNow environment. This behavior is the intended purpose of the skill for ServiceNow development and administration. The scripts provided are standard implementations of ServiceNow ITIL processes using the official GlideRecord and GlideAggregate APIs.
Audit Metadata