Agent skill · data analytics · delphine-l
jupyter-notebook-analysis
Best practices for creating comprehensive Jupyter notebook data analyses with statistical rigor, outlier handling, and publication-quality visualizations. Includes Claude API image size helpers.
Why this skill is useful
Provides specific statistical methods and image handling scripts for Jupyter notebooks that enhance data analysis capabilities.
What it needs
About 5k tokens when loaded. Last updated 2026-05-14. 17 stars on the source repository.
What this skill does
Jupyter Notebook Analysis Patterns Expert knowledge for creating comprehensive, statistically rigorous Jupyter notebook analyses. When to Use This Skill Creating multi-cell Jupyter notebooks for data analysis Adding correlation analyses with statistical testing Implementing outlier removal strategies Building series of related visualizations (10+ figures) Analyzing large datasets with multiple characteristics Building data update/enrichment notebooks with multi-source merging Generating figures for sharing with Claude or other AI tools Important: Image Size Constraints When generating images to share with Claude, images must not exceed 8000 pixels in either dimension. Add this helper to your notebook imports: For complete image size guidance, see the data-visualization skill. Core Notebook Patterns Data Update/Enrichment Notebooks Use structured notebook patterns for multi-source data merging and enrichment. Key principles: 1. Configuration section at top with safety defaults (ENABLEAWSFETCH = False, TESTMODE = True) 2. Composite keys for complex merge uniqueness requirements 3. Conflict resolution with configurable strategy (NEW vs OLD priority) 4. Idempotent column addition -- check if columns exist before adding 5. Enrichment tracking -- count what was actually saved, not just fetched 6. Two-stage file workflow -- input file -> distinct output file (never in-place) For detailed patterns including data update, enrichment, and AWS GenomeArk workflows, see notebook-patterns.md. Notebook Editing Always use NotebookEdit tool for .ipynb file modifications -- never the Edit tool (corrupts JSON structure). Three modes: replace (update cell content), insert (add new cell after target), delete (remove cell). …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills delphine-l/jupyter-notebook