trauma-informed-practice-designer
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill contains only instructional text and prompt templates. No executable scripts, network activity, or sensitive data access patterns were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill processes multiple user-input fields which are interpolated into the system prompt, creating a potential surface for indirect injection. However, since the skill has no dangerous capabilities (no tools, network, or file access), this surface cannot be exploited to perform malicious actions on the system.
- Ingestion points: SKILL.md fields including classroom_situation, known_context, and current_responses.
- Boundary markers: Bold text labels are used as delimiters; no explicit instructions for the model to disregard embedded commands are present.
- Capability inventory: No subprocess calls, executable code, file writes, or network operations were found in the skill.
- Sanitization: No input sanitization or validation logic is implemented.
Audit Metadata