skills/aipoch/medical-research-skills/generic-phenotype-scoring-research-planner/Gen Agent Trust Hub
generic-phenotype-scoring-research-planner
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of Markdown-based instruction files and reference templates. It does not include any scripts, binaries, or executable logic.
- [DATA_EXFILTRATION]: No patterns for network calls, credential harvesting, or access to sensitive file paths were detected. The skill correctly references its source on GitHub as standard vendor documentation.
- [INDIRECT_PROMPT_INJECTION]: The skill manages user research contexts using strict validation boundaries. 1. Ingestion points: User research contexts provided as input to the agent. 2. Boundary markers: Explicit 'Input Validation' and 'Out-of-scope' sections. 3. Capability inventory: None (no code execution or network writes). 4. Sanitization: Disease-context filtering and scope-specific redirection logic.
- [REMOTE_CODE_EXECUTION]: References to bioinformatics tools like limma, Seurat, and GSVA are limited to methodological recommendations in the resulting text output and do not involve package installation or shell execution by the skill.
Audit Metadata