kud-knowledge-type-mapper
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes several user-controllable input fields such as curriculum_input and existing_learning_targets which are interpolated directly into the model's instructions. While this creates a surface for prompt injection, the risk is mitigated by the lack of any dangerous capabilities or tools.
- Ingestion points: Variables in SKILL.md including {{curriculum_input}}, {{learner_stage}}, {{knowledge_architecture_output}}, {{existing_learning_targets}}, {{competency_framework}}, and {{programme_purpose}}.
- Boundary markers: No explicit delimiters (like XML tags or unique markers) are used to wrap these inputs.
- Capability inventory: The skill definition does not include any tools or scripts for network access, file writing, or command execution.
- Sanitization: No explicit sanitization or instructions for the model to ignore embedded commands within the user data are present.
Audit Metadata