duplicate-detection
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and analyzes user-generated content from ServiceNow tables, which is an inherent risk surface for indirect instructions. 1. Ingestion points: Steps 1 through 4 utilize the SN-Query-Table tool and REST endpoints to fetch data from kb_knowledge and kb_feedback tables, specifically accessing short_description, text, and comments fields. 2. Boundary markers: The procedure does not specify the use of delimiters or instructions to ignore potential commands embedded within the article text. 3. Capability inventory: The skill environment allows for record modification via SN-Update-Record and network interaction through the Bash tool. 4. Sanitization: There is no mention of sanitizing or validating the article content before it is processed by the agent.
Audit Metadata