tooluniverse-target-research
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses a 'Report-first' approach to create and progressively update a markdown report file based on gathered scientific data. This is a standard and safe practice for organizing complex tool outputs.
- [SAFE]: The skill interacts with numerous well-known scientific services and databases, including UniProt, Ensembl, OpenTargets, Pharos, and PubMed. These are established repositories for biological and medical data.
- [SAFE]: There is no evidence of prompt injection, obfuscation, or data exfiltration. Instructions regarding 'CRITICAL' or 'MANDATORY' steps refer to research methodology and report structure, not bypassing safety guidelines.
- [SAFE]: Network operations are restricted to queries against official scientific APIs via the provided tool interface (
tu.tools). No unauthorized connections to unknown third-party domains or command-and-control servers were detected. - [SAFE]: The skill handles identifier resolution and tool parameter verification (e.g., checking if a tool uses
idoruniprot_id) to ensure reliable data retrieval and prevent runtime errors.
Audit Metadata