discovery-patterns

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill provides patterns for querying and modifying the discovery_credentials table. While intended for administrative setup, this provides a pathway for interacting with sensitive authentication data within the ServiceNow environment.\n- [COMMAND_EXECUTION]: The skill contains templates for ServiceNow "Probes" that execute arbitrary shell commands on MID Servers using the OperatingSystemCommand API (e.g., cat /opt/myapp/version.txt).\n- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates a pattern where data collected from external network devices via probes is directly processed and stored in the CMDB without sanitization.\n
  • Ingestion points: The discovery_sensors_script template ingests result.output from external probes.\n
  • Boundary markers: None.\n
  • Capability inventory: The skill has the ability to write to the database (ci.update()) and execute scripts (snow_execute_script).\n
  • Sanitization: No input validation or escaping is present in the provided sensor script example.\n- [DYNAMIC_EXECUTION]: The skill utilizes the snow_execute_script tool and provides instructions for creating and executing dynamic scripts on both the ServiceNow instance and MID Servers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 07:16 PM
Security Audit — agent-trust-hub — discovery-patterns