cuequivariance
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The documentation focuses on providing guidance for using the 'cuequivariance' library, specifically for defining custom irreducible representations (Irreps) and constructing segmented polynomials. All code examples demonstrate standard Python practices for scientific computing.
- [SAFE]: The library dependencies identified (numpy) are standard for mathematical operations in this domain. No suspicious external downloads or remote code execution patterns were found.
- [SAFE]: The skill uses clear Einstein notation for tensor contractions, which is typical for this type of software and does not pose a security risk in this context.
- [SAFE]: The evaluation suite in 'evals.json' correctly models expected agent behavior, testing for both positive domain-specific capabilities and negative constraints against unrelated queries.
Audit Metadata