ladder-of-inference-reflection
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input that is directly interpolated into the system prompt.
- Ingestion points: The situation, context, student_level, current_interpretation, and desired_use fields in the SKILL.md prompt.
- Boundary markers: Absent. The inputs are placed directly into the prompt template (e.g., Situation: {{situation}}) without delimiters like XML tags or explicit instruction separators.
- Capability inventory: The skill has no file system access, no network operations, and no command execution capabilities across the analyzed file.
- Sanitization: No evidence of input sanitization or filtering to prevent instruction injection.
Audit Metadata