figure-table-quality

Figure & Table Quality Audit

Pipeline position: Phase 2.5 (between Grounding/Polish and Submission). Runs after figure-rhetoric (content) and before arxiv-preflight (compliance). See /manuscript-pipeline for full execution order.

Purpose

Verify that every figure and table in a manuscript renders at readable size in the compiled PDF. figure-rhetoric checks whether figures communicate the right message. This skill checks whether the reader can physically read them.

Execution

Step 1 — Build the display-scale map

For every \includegraphics in the .tex source:

1. Extract the display width (e.g., \textwidth, 0.7\textwidth, 0.55\textwidth)
2. Compute the effective display width in inches using the document geometry
3. Read the figure generation script to find the figsize for each figure
4. Compute: scale = display_width / figsize_width

Also extract from the generation script or config:

• Base font sizes: title, label, tick, legend, annotation
• Marker sizes, line widths
• Bar widths (for grouped bar charts)

Step 2 — Per-figure audit (all 9 checks)

For each figure, compute effective values at display scale and check:

2a. Font size at render

effective_font = script_font × scale

Element	Minimum	Warning
Axis title	7pt	8pt
Tick labels	6pt	7pt
Legend text	6pt	7pt
Annotations	6pt	7pt
Panel titles	7pt	8pt

FAIL if any element falls below minimum. WARN if any element falls below warning threshold.

2b. Label collision

Check for overlapping text in:

• X-axis tick labels (especially with rotation < 45° and > 3 labels)
• Y-axis tick labels (long text strings)
• Data annotations near each other
• Legend entries overlapping data

For rotated labels: compute horizontal footprint as len(label) × char_width × cos(rotation). If footprint > tick spacing, FAIL.

2c. Marker and line visibility

effective_marker = script_marker × scale
effective_linewidth = script_linewidth × scale

• Markers below 4pt effective: WARN
• Line widths below 0.5pt effective: WARN