概要
The Notebook Data Consistency skill provides a rigorous framework for maintaining data integrity within complex Jupyter notebooks, particularly those handling large datasets like single-cell proteomics. It guides Claude to trace variable dependency chains, verify data sources through grep-based reference scanning, and implement proactive shape assertions. By enforcing best practices for AnnData subsetting and threshold synchronization, this skill eliminates the common 'shape mismatch' errors that occur when updating pipelines or changing input datasets.
