The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external ServiceNow tables to generate reports and summaries.
Ingestion points: The skill retrieves data from the sn_grc_issue, sn_grc_risk, and sn_compliance_control tables, specifically reading the description and short_description fields (SKILL.md, Steps 1-5).
Boundary markers: The instructions lack explicit delimiters or instructions to the agent to disregard potential instructions embedded within the retrieved GRC records.
Capability inventory: The skill can execute server-side logic via SN-Execute-Background-Script and perform network operations via Bash (ServiceNow REST API).
Sanitization: No sanitization or filtering is applied to the external data before it is processed by the agent.
[COMMAND_EXECUTION]: The skill utilizes the SN-Execute-Background-Script tool to run ServiceNow GlideScript. While the scripts provided in the skill are designed for legitimate GRC calculations and reporting logic, the tool itself provides high-privilege access to the ServiceNow instance.
[DYNAMIC_EXECUTION]: The skill performs dynamic execution of JavaScript logic via the SN-Execute-Background-Script tool to process complex aggregations, metrics, and aging analysis at runtime (SKILL.md, Steps 4, 5, and 6).

issue-summarization