Chart Prompts

1,337 ways to dress a line or bar chart, 669 light and 668 dark. Open any chart to see it full size and copy a prompt that gives your own chart the same look.

Every chart uses illustrative data. The prompts work with any coding agent and any charting library; they describe colors, type, spacing and marks precisely enough to reproduce each look.

← More from Emergent Mind Labs

Stability Study bar chart, light theme: Clean QC report with paired condition bars, a red spec limit and OOS flags.

Stability Study

LightBar

Clean QC report with paired condition bars, a red spec limit and OOS flags.

  • Slate blue#4F6D8F
  • Amber#D98E2E
  • Spec red#C62F2B
  • Ink#1F2A36

Prompt for your coding agent

Paste it with your chart's code. It restyles your data to match this chart.

Restyle my chart in the "Stability Study" style described below. Keep my data, chart type, title meaning and units; change only the presentation. Use the charting library this project already uses (plain SVG if there is none). If the library can't express a detail, draw it in a custom layer instead of dropping it. Measurements assume a 960 x 600 chart; scale them proportionally.

Mood: a quality control stability summary. Clean white, sober and regulatory: two storage conditions side by side, one red specification line and an unmistakable flag on any failing result.

Colors (use these exact values):
- Page: #FFFFFF. Ink: #1F2A36. Secondary text: #6A7480. Gridlines: #E7EAEE.
- Condition 1 (long term): #4F6D8F (slate blue). Condition 2 (accelerated): #D98E2E (amber).
- Specification and failures: #C62F2B (red), used only for the limit line, its label, the outline and flag of failing marks and their values.
- Not-tested placeholders: 1px dashed (3 3) #C9CED6.

Typography (Google Fonts): Questrial (a single 400 weight) for everything.
- Eyebrow: 12.5px uppercase, 1.6px letter-spacing, #6A7480, after a 4 x 22 px slate bar. Title: 29px #1F2A36, stating the finding. Subtitle: 15px #6A7480 with measure, unit and what the limit means.
- Legend 13.5px; tick labels 13px; value labels 12.5px; category labels 14px; notes and source 12px.

Layout:
- 56px margins. Eyebrow at y 57, title at y 100, subtitle at y 128, a legend row (13px squares, radius 2) at y 160. Plot from x 96 to x 896, baseline at y 470.
- Footer: an abbreviations line ("n/t = not tested. OOS = out of specification.") above "Source: ..." at left, a document reference at right.

Axes and grid:
- Y axis from zero with percent tick labels every 2 points and 1px #E7EAEE gridlines; no y axis line. A 1.2px #1F2A36 baseline. Timepoint labels 24px under the baseline ("Initial", "3 mo", ...).
- The specification limit is a 1.6px #C62F2B dashed (7 4) line across the plot, labelled at the right end above the line on a white knockout ("Specification limit 5.0%").

Marks:
- Bar chart: grouped vertical bars, 34px wide with a 4px gap inside the pair and 2px top corners. Value labels 7px above each bar with a 4px white halo so they read across the limit line. A bar that breaks the limit keeps its fill but gets a 2px #C62F2B inner outline, its value in red to the left of a 1.4px red flagpole, and a 34 x 14 red pennant (notched right edge) reading "OOS" in 10.5px white. Timepoints with no result show a dashed empty placeholder with "n/t" above it.
- Line chart: two 2.5px lines in the condition colors with 4px dots at each timepoint; the same red dashed limit; the first point past the limit becomes a 6px red-ringed dot with the OOS pennant; untested timepoints end the line (no interpolation) with a short note.
- If the underlying measure is a narrow percentage (for example purity from 90 to 100%), plot its complement (impurities or loss from initial) from zero so bar lengths stay honest, and say so in the subtitle.

Avoid: truncated y axes, gradients, shadows, more than one red element per failure, and decorative icons.

When you're done, render the chart and check that bars start at zero, only failing marks are red-outlined and flagged, no value label is struck through by the limit line, and the title's timepoint matches the first failure.