Review AI/ML experiment Python code for correctness and reproducibility.
npx skills add https://github.com/schmidtkk/skills --skill ml-code-review
Please help me install this Agent Skill. Skill: ml-code-review Source: https://github.com/schmidtkk/skills/tree/main/ml-code-review Command: npx skills add https://github.com/schmidtkk/skills --skill ml-code-review
Reviews AI/ML experiment Python code for correctness, reproducibility, and best practices across PyTorch, TensorFlow, and JAX. It helps identify issues in training loops, data handling, and architectural choices that can affect results and reproducibility.
Provide a comprehensive ML code review for a given training script to ensure correctness, reproducibility, and adherence to best practices.
High-intent search queries and answers about installing and using this skill.
To get actionable fixes for issues found during an ML experiment code review, this Skill generates structured reports detailing severity levels, concrete remediation steps, and example invocations to speed up debugging and auditing.
To review ML training code for reproducibility and correctness, this Skill analyzes Python scripts to verify seed setting, determinism, training loop logic, and data handling, providing structured reports with actionable fixes.
Yes, this ML code review tool supports PyTorch, TensorFlow, and JAX codebases, applying domain-aware checks for training loop correctness, reproducibility, and architectural best practices across these frameworks.
Yes, you can check diffusion model, Transformer, GAN, and VAE implementations for architectural best practices, receiving actionable remediation steps and structured reports to ensure correctness and adherence to established guidelines.
A machine learning code review checks data loaders and experiment scripts for data handling correctness, reproducibility issues like missing seeds, and training loop errors, delivering a structured report with severity levels and concrete fixes.
To get actionable fixes for issues found during an ML experiment code review, this Skill generates structured reports detailing severity levels, concrete remediation steps, and example invocations to speed up debugging and auditing.
raw-labsCreate, validate, and package Claude Skills with Python scripts.
SALTYf1SHTranslate research protocols into runnable PyTorch or TensorFlow training pipelines.
trymiraiValidates inference outputs by comparing them with reference results via Rust cargo test harness.
ahiendzLint and statically analyze Node.js/TypeScript and Python codebases.
