Does it support industry-specific standards?

Yes, it can cite and align with authoritative standards such as NASA-STD-7009 and ASME VVUQ for verification and validation.

When should I use the /math-review command?

Use it whenever you are implementing or modifying complex algorithms, statistical routines, or any code where precision and mathematical accuracy are critical.

Can this skill help with machine learning models?

Yes, it is specifically designed to review ML/AI model implementations, ensuring that the underlying mathematical logic is sound and the implementation is stable.

What does the Math Review skill do?

It performs a deep-dive analysis of mathematical code to check for derivation errors, numerical instability, and non-compliance with scientific standards.

Mathematical Algorithm Review

Name: Mathematical Algorithm Review
Author: athola

byathola

•

Security & Testing

Performs intensive mathematical analysis to ensure numerical stability, algorithmic correctness, and alignment with authoritative scientific standards.

The Math Review skill provides a rigorous framework for auditing math-heavy codebases, including machine learning models, scientific computing scripts, and financial algorithms. It automates the process of mapping requirements to mathematical invariants, verifying derivations through Computer Algebra Systems (CAS), and assessing numerical stability to prevent issues like precision loss or overflow. By integrating industry standards such as NASA-STD-7009, this skill ensures that complex calculations are not only correct but also robust and well-documented for safety-critical environments.

Key Features

01Requirements-to-invariant mapping and coverage analysis

02Progressive loading for scalable analysis depth

03Numerical stability assessment including precision and conditioning

04Evidence logging with benchmark and notebook execution

05Automated derivation verification against authoritative standards

0610 GitHub stars

Use Cases

01Validating safety-critical or financial calculations for production readiness

02Auditing scientific computing code for numerical integration and optimization

03Verifying machine learning model implementations and probabilistic logic

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add athola/claude-night-market math-review

For use in Claude.ai and ChatGPT

Download Skill