icd10-codes
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses the
@sequoiaport/codesNode.js package and makes network requests toapi.sequoiacodes.com. Both are verified as official resources belonging to the skill author, SequoiaPort. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it retrieves external medical data for agent processing.
- Ingestion points: Clinical descriptions and code metadata retrieved from
api.sequoiacodes.comas described inSKILL.md. - Boundary markers: None identified within the skill instructions.
- Capability inventory: The skill is limited to performing authenticated network API calls for data retrieval.
- Sanitization: No explicit sanitization of the retrieved medical descriptions is provided in the documentation.
Audit Metadata