The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from ServiceNow tables that may contain untrusted user-generated content, which could be used to influence the agent's behavior.
Ingestion points: The skill fetches fields such as short_description, description, work_notes (from sys_journal_field), root_cause, and contributing_factors from sn_risk_event records.
Boundary markers: No specific delimiters or instructions to ignore embedded prompts are provided when interpolating this data into the summarization context.
Capability inventory: The skill utilizes SN-Execute-Background-Script (for server-side JavaScript execution on ServiceNow), Bash, and various REST API tools to interact with the environment.
Sanitization: There is no evidence of sanitization, escaping, or filtering of the ingested record content before it is processed by the agent.
[DYNAMIC_EXECUTION]: The skill relies on generating and executing server-side scripts to perform data aggregation and narrative generation.
Evidence: Steps 6 and 7 provide JavaScript templates for the SN-Execute-Background-Script tool, using ServiceNow-specific APIs like GlideRecord, GlideAggregate, and gs.dateDiff. The agent is expected to populate these templates with runtime values (e.g., EVENT_SYS_ID) and execute them.

risk-event-summarization