quality-enriched-prompting
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted prompt and context data to perform quality assessments and improvements via an external LLM. Ingestion points: The prompt and context arguments in the llm_quality_eval and llm_quality_improve functions in skill.md. Boundary markers: Absent. The system messages do not wrap input strings in delimiters or instruct the model to ignore instructions within the inputs. Capability inventory: Interaction with the OpenAI API for chat completions. Sanitization: Absent. This creates a surface where the prompts being analyzed could potentially influence the behavior of the evaluation model, which is a standard risk for prompt-analysis tools.
Audit Metadata