The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes task record data which is a potential source of untrusted instructions.
Ingestion points: Reads data from task, sc_task, and task_sla tables, including potentially user-provided fields like short_description (SKILL.md).
Boundary markers: Absent; there are no specific instructions for the agent to treat record content as untrusted or to ignore embedded instructions.
Capability inventory: The skill has access to SN-Execute-Background-Script for server-side JavaScript execution and SN-Add-Work-Notes for writing back to records (SKILL.md).
Sanitization: No explicit sanitization or filtering of external record content is described before the data is processed or included in reports.
[DYNAMIC_EXECUTION]: The skill generates and executes scripts within the ServiceNow platform environment.
Execution method: Utilizes the SN-Execute-Background-Script tool to perform complex data aggregations and calculations (SKILL.md).
Implementation: The skill provides several hardcoded JavaScript templates for analyzing task volume, bottlenecks, and trends. While the logic is predefined in the skill, it represents a dynamic code execution surface on the target platform.

task-analysis