conversation-evaluator
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted user-generated text from ServiceNow conversation logs.
- Ingestion points: User messages are retrieved from the sys_cs_message table in Step 2 of the procedure.
- Boundary markers: The instructions do not specify the use of boundary markers or clear delimiters to isolate the untrusted conversation content from the agent's evaluation instructions.
- Capability inventory: The skill has access to the Bash native tool and several ServiceNow modification tools (SN-Update-Record and SN-Add-Work-Notes).
- Sanitization: There is no evidence of message sanitization or filtering applied to the retrieved conversation message bodies before they are presented to the AI for evaluation.
Audit Metadata