intake-interview
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill utilizes an 'Interactive Mode' where the agent collects patient data step-by-step and provides a clinical synthesis. This process is susceptible to indirect prompt injection if a user provides instructions rather than patient data to influence the summary.
- Ingestion points: User responses during the interview interaction in
SKILL.md. - Boundary markers: Absent; there are no specific instructions for the agent to treat input as literal text or ignore potentially malicious instructions within the input.
- Capability inventory: None; the skill is entirely text-based and does not utilize any tools for network access, file system operations, or command execution.
- Sanitization: Absent; the skill does not implement any validation or filtering for the data provided by the user.
Audit Metadata