notebook-debugger

What problem does it solve?

This skill guides users to diagnose and fix failing Jupyter notebooks by applying a formal, repeatable troubleshooting workflow that covers kernel crashes, import errors, and memory issues.

Core Features & Use Cases

• Structured troubleshooting workflow: Follows Diagnose → Isolate → Fix → Verify → Document phases to resolve notebook failures.
• Environment and kernel hygiene: Checks for correct Python environments, registered Jupyter kernels, and reproducible dependencies to ensure notebooks run in stable contexts.
• Memory and performance diagnostics: Provides guidance on memory profiling, chunking, and memory-safe data handling to prevent crashes on large datasets.
• Reproducibility and documentation: Emphasizes end-to-end execution on fresh kernels and produces setup guidance to support reproducibility across machines.

Quick Start

Open a failing notebook and follow the diagnostic workflow: restart the kernel, run all cells in displayed order, perform memory and import checks, apply the recommended fixes (reorganize cells, add dependency checks, enforce execution order, and document environment), then verify reproducibility with a fresh environment by performing a full Restart & Run All.