output-validation
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides comprehensive instructions for mitigating indirect prompt injection risks by validating and sanitizing data before it is processed by downstream tools or external sinks.
- Ingestion points: The skill focuses on processing model-generated outputs which may be influenced by untrusted external data.
- Boundary markers: It advocates for the use of strict schema contracts (Pydantic, Zod, JSON Schema) to define clear boundaries between the LLM and the application layer.
- Capability inventory: No executable scripts are included with this skill; it is instructional in nature.
- Sanitization: It recommends PII masking (using hybrid Regex/NER approaches like Microsoft Presidio) and strict schema validation to ensure data conforms to expected structures before any side-effecting actions occur.
- [EXTERNAL_DOWNLOADS]: The instructions mention well-known and reputable industry tools such as Microsoft Presidio, Pydantic, and Zod. These references are for implementing the defensive strategies described in the skill.
Audit Metadata