matplotlib
zLanqing/codex-claude-academic-skills / scientific-toolkit-skill/references/scientific-skills/matplotlib
gives you control over each plot element , which improves: custom charts, ready for publication
Details
- judge’s score
- 7/10
- place
- outside the packs
- publisher
- zLanqing
- status
- new
- pinned commit
- 7ed6377 on GitHub (new tab)
- skill’s last commit
- May 14, 2026
- repo stars
- 5,118
- main type
- Data
Judgement
Precise trigger and clear scope. Content mostly known to Claude, with a few concrete examples. Technical defects: boxplot(labels=) deprecated, savefig(optimize=) inaccurate, private API _rebuild().
reservations
- The api_reference, common_issues, plot_types and styling_guide annexes are truncated at 5,000 characters: the end could not be verified.
- Deprecated or inaccurate code: ax.boxplot(labels=) replaced by tick_labels since matplotlib 3.9, and savefig(optimize=True) is not a valid argument.
- Relies on the private API matplotlib.font_manager._rebuild(), which may break between versions.
- Mostly generic content (basic chart types), with little know-how specific to a maintainer.
- Triggering
- 2/2
- Added value
- 1/2
- Precision
- 2/2
- Scope
- 2/2
- Quality
- 0/2
Alerts
1 alert, by category
Static scan
- high risk
scripts run by the SKILL.md
scripts/plot_template.py, scripts/style_configurator.py
Original description
Low-level plotting library for full customization. Use when you need fine-grained control over every plot element, creating novel plot types, or integrating with specific scientific workflows. Export to PNG/PDF/SVG for publication. For quick statistical plots use seaborn; for interactive plots use p…
added on Oct 19, 2026 · license MIT