generate-plot

Generate a matplotlib chart/visualization from a natural-language description.

Workflow

1. •Parse the user's chart description from $ARGUMENTS
2. •Determine the appropriate chart type (bar, line, pie, scatter, etc.)
3. •Write a Python script using matplotlib to generate the chart
4. •Execute the script to produce the output file (PNG by default)
5. •Report the output file path to the user

Chart Type Selection

• •Bar chart: Comparing discrete categories or items
• •Horizontal bar: Comparing categories with long labels
• •Stacked bar: Showing composition across categories
• •Grouped bar: Comparing multiple series across categories
• •Line chart: Showing trends over time or continuous data
• •Pie/Donut chart: Showing proportions of a whole (use sparingly, max 7 segments)
• •Scatter plot: Showing correlation between two variables
• •Area chart: Emphasizing volume/magnitude over time
• •Heatmap: Showing patterns in matrix data

Script Template

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt
import matplotlib.ticker as ticker

# Azure color palette
AZURE_COLORS = [
    "#0078D4",  # Azure Blue (primary)
    "#50E6FF",  # Light Blue
    "#F25022",  # Red
    "#7FBA00",  # Green
    "#FFB900",  # Yellow
    "#737373",  # Gray
    "#B4009E",  # Purple
    "#00B7C3",  # Teal
    "#D83B01",  # Dark Orange
    "#005B70",  # Dark Teal
]

# Professional styling
plt.rcParams.update({
    "figure.facecolor": "white",
    "axes.facecolor": "white",
    "axes.grid": True,
    "axes.spines.top": False,
    "axes.spines.right": False,
    "grid.alpha": 0.3,
    "font.family": "sans-serif",
    "font.size": 11,
    "figure.dpi": 150,
})

fig, ax = plt.subplots(figsize=(10, 6))

# ... chart-specific code ...

plt.tight_layout()
plt.savefig("output.png", bbox_inches="tight", dpi=150)
print("Chart saved to output.png")

Style Guidelines

• •Always use matplotlib.use("Agg") before importing pyplot (no display needed)
• •Use the Azure color palette defined above
• •Remove top and right spines for a cleaner look
• •Add subtle grid lines (alpha=0.3) on the value axis
• •Use tight_layout() and bbox_inches="tight" when saving
• •Default figure size: 10x6 inches at 150 DPI
• •Add clear title, axis labels, and legend where appropriate
• •For pie charts, use a donut style (wedgeprops with width=0.6) and add percentage labels
• •Rotate x-axis labels if they overlap
• •Use commas in large numbers via ticker.FuncFormatter

Output

• •Default format: PNG at 150 DPI
• •If the user requests SVG, use plt.savefig("output.svg", format="svg")
• •Use a descriptive filename based on the chart content (e.g., azure-cost-comparison.png)
• •Always print the output path after saving

Data Handling

• •If the user provides explicit data, use it directly
• •If the user describes a concept without data, generate realistic sample/placeholder data and clearly label it as illustrative
• •For Azure cost data, use realistic price ranges from public Azure pricing
• •Parse inline data formats: CSV snippets, JSON, markdown tables, or plain lists