rw-research-data
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a security validation script (
scripts/self_check.py) that scans all skill files for prohibited strings representing local sensitive paths (e.g.,/Users/,private-workspace/,~/.claude), ensuring the skill remains portable and does not leak environment-specific information.- [SAFE]: Instructions withinSKILL.mdandreferences/method.mdinclude explicit safety constraints and 'STOP' conditions, directing the agent not to bypass login portals, embargoes, or data usage agreements (DUAs) when verifying research material availability.- [EXTERNAL_DOWNLOADS]: The skill references established research and metadata standards from organizations like GO-FAIR, DataCite, ICMJE, and COPE. These links are provided for verifying current regulatory requirements and do not involve automated execution or untrusted payloads.
Audit Metadata