skills/aipoch/medical-research-skills/nhanes-clinical-retrospective-biomarker-research-planner/Gen Agent Trust Hub
nhanes-clinical-retrospective-biomarker-research-planner
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes a file (
eval_report_nhanes-clinical-retrospective-biomarker-research-planner_result.json) containing self-referential evaluation data that explicitly claims the skill is safe and secure. This targets the analyzer itself to influence the audit outcome. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through user-provided study parameters. Ingestion points: User-provided disease and biomarker definitions in SKILL.md. Boundary markers: Absent; the instructions do not specify delimiters to separate untrusted user data from agent instructions. Capability inventory: The skill is limited to text generation and does not have file-write, network-access, or subprocess capabilities across its modules. Sanitization: Absent; there is no mention of filtering or escaping user-provided input.
Audit Metadata