Content
52%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a well-organized, code-rich overview with genuinely useful best-practice guidance (colormaps, DPI, layout, accessibility), but it spends much of its ~350 lines teaching matplotlib fundamentals Claude already knows, and its progressive-disclosure architecture is illusory — all six referenced bundle files are missing. Tightening the tutorial material and either shipping or removing the dangling references would substantially improve it.
Suggestions
Create the referenced bundle files (references/plot_types.md, styling_guide.md, api_reference.md, common_issues.md, scripts/plot_template.py, style_configurator.py) or remove the references and the Quick Reference Scripts / Detailed References sections that point to them.
Cut sections that restate matplotlib knowledge Claude already has — the Figure/Axes/Artist/Axis hierarchy, the pyplot-vs-OO tutorial, and basic gotchas — and keep only the curated judgment calls (colormap selection, DPI by medium, constrained_layout, blitting).
Make code examples self-contained and copy-paste runnable by defining the data they use (e.g., generate x/y with numpy in the subplot and 3D examples) instead of referencing undefined variables.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Noticeably verbose with several padded sections that restate matplotlib knowledge Claude already has: the Figure/Axes/Artist/Axis hierarchy tutorial, the pyplot-vs-OO two-interfaces walkthrough, and gotchas like 'figsize is in inches, not pixels'. Valuable curated guidance (colormap choice, DPI by medium, constrained_layout) is diluted by textbook explanation, fitting the 'several unnecessary explanations' anchor rather than the mostly-efficient one. | 2 / 5 |
Actionability | Abundant concrete, runnable code across plot types, styling, saving, and 3D. Minor gaps keep it below fully copy-paste ready: several snippets use undefined variables (x, y1, categories, values, data, X/Y/Z, my_data) in the subplots, plot-types, and 3D sections. | 4 / 5 |
Workflow Clarity | Common workflows are clearly sequenced with concrete steps (create figure/axes, plot, customize, save, show) and the code-organization section models a good reusable structure. No validation checkpoints are present, but there are no destructive or batch operations, so the workflow cap does not apply; only the implicitness of checkpoints separates it from a 5. | 4 / 5 |
Progressive Disclosure | The body is well-sectioned and references are clearly signaled, but every referenced bundle path is broken: references/plot_types.md, references/styling_guide.md, references/api_reference.md, references/common_issues.md, scripts/plot_template.py, and scripts/style_configurator.py do not exist. Scoring against the actual bundle structure, navigation leads nowhere, which is a worse failure than buried references. | 2 / 5 |
Total | 12 / 20 Passed |