tooluniverse-spatial-omics-analysis
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a logical and well-documented workflow for scientific data analysis using structured tool calls to reputable bioinformatics APIs.
- [DATA_EXFILTRATION]: No unauthorized data access or exfiltration patterns were found. All network activity is directed toward established biological databases (e.g., STRING, Human Protein Atlas, OpenTargets) as part of the intended functionality.
- [PROMPT_INJECTION]: The instructions are focused on analysis phases and report generation; no attempts to bypass safety filters or override system instructions were identified.
- [COMMAND_EXECUTION]: The skill does not invoke shell commands or execute arbitrary scripts. Its operations are limited to data processing and information retrieval via predefined tools.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided gene lists and experimental metadata, it does so using specialized biological tools that do not present a path for escalating privileges or performing harmful actions.
Audit Metadata