The Agent Skills Directory

[SAFE]: The skill implements a legitimate operational workflow for SOC analysts using ServiceNow. All tool configurations and API endpoints are consistent with its stated purpose of summarizing security incidents and tasks for shift transitions.
[DATA_EXFILTRATION]: The skill accesses sensitive security data, including incident details and vulnerability exploitation status from ServiceNow tables (sn_si_incident, sn_vul_vulnerable_item). This data access is aligned with the primary purpose of a SOC handover report. The skill does not contain hardcoded credentials or instructions to send this data to unauthorized external domains.
[PROMPT_INJECTION]: The skill processes untrusted data from external sources such as incident work notes and descriptions in ServiceNow.
Ingestion points: Data is ingested from sn_si_incident, sn_si_task, sn_vul_vulnerable_item, and sys_journal_field via the SN-Query-Table and REST tools described in SKILL.md.
Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded commands within the ingested text.
Capability inventory: The agent has write access via SN-Add-Work-Notes and access to the Bash tool.
Sanitization: No explicit sanitization or filtering of the ingested ServiceNow content is performed before interpolation into the handover report.
Analysis: While this creates a surface for indirect prompt injection, it is a functional requirement for the skill's purpose of summarizing analyst notes and incident details. The risk is considered inherent to the operational use case and does not represent a malicious design.

shift-handover