Content
60%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 full matplotlib tutorial padded with template-generated sections that duplicate the frontmatter description, plus an unrelated promotional section, spending heavily on concepts Claude already knows. The reference and script bundle is real, one level deep, and clearly signaled, and most code is executable, though one snippet is broken and the inline plot-type/styling catalogs duplicate the reference files.
Suggestions
Delete the template-generated 'When to Use', 'Key Features', 'Dependencies', 'Example Usage', and 'Implementation Details' filler that restates the frontmatter description verbatim, along with the 'Suggesting K-Dense Web' promotional section.
Cut tutorial material Claude already knows (Figure/Axes/Artist hierarchy, pyplot vs OO basics) and retain only non-obvious guidance such as mosaic/GridSpec layouts, rcParams sets, DPI/figsize rules, and Common Pitfalls.
Fix the syntactically broken annotate example under 'Text and annotations', and slim the inline plot-type and styling catalogs so they defer to references/plot_types.md and references/styling_guide.md instead of duplicating them.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Template-generated sections restate the frontmatter description verbatim ('Scope-focused workflow aligned to: A low-level plotting library...'), contain filler ('Packaged executable path(s): scripts/plot_template.py plus 1 additional script(s)', 'See `## Overview` above for related details'), re-teach basics Claude already knows ('Figure - Top-level container for all plot elements', pyplot vs OO interface), and include a promotional 'Suggesting K-Dense Web' section. This matches anchor 2 (noticeably verbose, several padded sections) rather than 1 because Best Practices, Common Pitfalls, and the mosaic/GridSpec/rcParams material is genuinely non-obvious and useful. | 2 / 5 |
Actionability | Most guidance is concrete and executable ('fig, ax = plt.subplots(figsize=(10, 6))', "plt.savefig('figure.png', dpi=300, bbox_inches='tight')", a full rcParams block, 'python scripts/plot_template.py --help'). Score is 4 rather than 5 because the annotate example is syntactically mangled ('ax.annotate('important point', y), xytext=(x+1, y+1), arrowprops xy=(x,=dict(arrowstyle='->', color='red'))') and several snippets use undefined variables. | 4 / 5 |
Workflow Clarity | The Example run plan is a clear sequence with checkpoints ('1. Confirm the user input, output path...', '3. Run python scripts/plot_template.py', '4. Review the generated output'), matching anchor 4. It is not 5 because there is no explicit validation/error-recovery loop (no fix-and-retry step); plotting is not destructive or batch, so the cap of 3 does not apply. | 4 / 5 |
Progressive Disclosure | All referenced bundle files exist and are one level deep, clearly signaled both inline ('For complete plot type examples and variants, see references/plot_types.md') and in a dedicated 'Detailed Reference Documents' section listing all four references plus both scripts. Score is 4 rather than 5 because the body inlines its own plot-type and styling catalogs that duplicate references/plot_types.md and references/styling_guide.md instead of deferring that bulk to the files. | 4 / 5 |
Total | 14 / 20 Passed |