skills/serac-labs/serac/csm-patterns/Gen Agent Trust Hub

csm-patterns

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The Portal Case Submission widget server script template ingests untrusted user data from the input object and interpolates it directly into ServiceNow case records. This creates a surface for indirect prompt injection where malicious instructions could be persisted in the database and later influence the behavior of agents or systems processing those records.\n
  • Ingestion points: The input.subject and input.description fields in the Portal Case Submission widget server script (SKILL.md).\n
  • Boundary markers: No boundary markers or 'ignore' instructions are present to delimit user-provided content.\n
  • Capability inventory: The skill utilizes tools such as snow_execute_script and snow_artifact_manage which allow for script execution and configuration changes in the ServiceNow environment.\n
  • Sanitization: No sanitization or validation logic is applied to the input before it is set on the case record.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 07:16 PM
Security Audit — agent-trust-hub — csm-patterns