scientific-method
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a reference guide for research methodology and does not contain any malicious code, prompt injections, or unauthorized network operations.
- [SAFE]: The provided Python templates and shell commands (e.g., git metadata extraction) are standard practices for computational reproducibility and version control, designed to help agents generate well-documented research code.
- [SAFE]: The reference files include security best practices, such as dependency pinning with hashes (SHA-256) and using deterministic Docker environments to prevent supply chain risks in research pipelines.
- [SAFE]: All external URLs point to well-known research repositories (OSF, AsPredicted) or official documentation (PyTorch, Wikipedia) and are used for educational or validation purposes.
Audit Metadata