mistral-sdk-patterns
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill provides patterns for processing untrusted external data (e.g., ticket classification), which constitutes an indirect prompt injection surface.
- Ingestion points: The
textparameter in theclassifyTicketfunction inSKILL.md. - Boundary markers: Employs the
json_schemaresponse format to constrain model output. - Capability inventory: The skill is scoped with
Read,Write, andEdittools. - Sanitization: Implements Zod schema validation to verify the integrity and structure of model-generated data at runtime.
Audit Metadata