depression-screening
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation, clinical decision trees, and behavioral instructions for an AI agent to conduct psychometric screening. All referenced files are local markdown documents.
- [NO_CODE]: No executable scripts (Python, JS, Shell), environment configurations, or dependency manifests were detected in the skill package.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an interactive administration workflow where the agent ingests user input (responses to questionnaire items). While this creates an entry point for untrusted data, the skill does not request or utilize any external tools, network access, or file-writing capabilities that could be leveraged for an exploit. The implementation includes specific safety guardrails (Item 9 protocol) to prioritize clinical safety.
- [DYNAMIC_CONTEXT_INJECTION]: No shell-interpolated placeholders or dynamic loading patterns were found in the instructions.
Audit Metadata