skills/freedomintelligence/openclaw-medical-skills/tooluniverse-gwas-study-explorer/Gen Agent Trust Hub
tooluniverse-gwas-study-explorer
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill primarily implements statistical methods for meta-analysis (Fixed-Effects and Random-Effects models) using standard Python libraries such as
mathandre. - [EXTERNAL_DOWNLOADS]: The documentation references the
tooluniversepackage, which is the host framework for the skill. Downloads and installations are performed through standard, well-known registries. - [DATA_EXFILTRATION]: The skill fetches scientific association data from public, trusted genomic databases (GWAS Catalog and Open Targets) via a tool-call abstraction (
tu.run). No access to sensitive user files, credentials, or arbitrary network domains was found. - [COMMAND_EXECUTION]: Shell command execution is limited to standard development and testing workflows (e.g.,
pytest) as described in the documentation, with no evidence of unauthorized or hidden command execution patterns. - [PROMPT_INJECTION]: The skill contains no instructions aimed at overriding agent behavior or bypassing safety filters. All instructional content is focused on GWAS methodology and data interpretation.
Audit Metadata