document-screening
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill processes inbound document data and extraction results, creating an indirect prompt injection surface where instructions embedded in documents could influence the agent's logic for routing and task creation.\n
- Ingestion points: Document data and extraction results are ingested from the sys_attachment and sn_doc_intelligence_extraction_result tables in ServiceNow (referenced in Step 1 and Step 3).\n
- Boundary markers: No explicit boundary markers or instructions to disregard embedded content are included in the skill definition.\n
- Capability inventory: The agent can perform write operations and modify records using SN-Create-Record, SN-Update-Record, and SN-Add-Work-Notes.\n
- Sanitization: No data sanitization or filtering steps are specified for the ingested document content before it is used to influence the workflow.\n- [COMMAND_EXECUTION]: The skill describes the use of Bash and curl to perform REST API calls to ServiceNow. This behavior is consistent with the skill's intended purpose of document triage and data management.\n- [EXTERNAL_DOWNLOADS]: The skill references official ServiceNow documentation and configuration guidelines hosted on their public GitHub repository for implementation support.
Audit Metadata