natural-language-queries
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external ServiceNow records (such as incident descriptions), which serves as an ingestion point for potentially untrusted content that could contain malicious instructions for the agent.
- Ingestion points: Records retrieved via
SN-Natural-Language-SearchandSN-Query-Tableas described in SKILL.md. - Boundary markers: The skill instructs the agent to inspect
unmatchedTextas a validation failure and emphasizes minimal field selection to reduce exposure. - Capability inventory: The skill environment includes tools for querying and reading ServiceNow table data.
- Sanitization: The documentation recommends manual verification of results and checking against known records, though no automated sanitization of the retrieved text fields is specified.
Audit Metadata