ai-expertise-interrogation-designer
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates untrusted user input directly into its core prompt template, creating a potential surface for indirect prompt injection.\n- Ingestion points: Variables defined in the input_schema, such as student_expertise_domain, student_level, and target_ai_tool, are interpolated verbatim into the prompt in SKILL.md.\n- Boundary markers: The template lacks delimiters (e.g., XML tags or block quotes) around the interpolated variables to isolate user-provided data from the agent's instructions.\n- Capability inventory: The skill is limited to text generation; it does not possess capabilities for network requests, file system access, or command execution.\n- Sanitization: No sanitization, validation, or escaping logic is implemented to ensure that user inputs do not contain malicious instructions intended to steer the agent's behavior.
Audit Metadata