Data analysis tools enable agents to create charts, visualizations, and analyze data using various formats and libraries.
Generates static chart images from Plotly configuration provided as JSON or YAML.
Parameters
The Plotly configuration as a JSON or YAML string. Must be a complete Plotly figure configuration.
The format of the config_content. Must be either "json" or "yaml".
The desired filename for the output image artifact.
The desired image format. Options: "png", "jpg", "svg", "pdf", "webp".
Usage Example
from solace_agent_mesh.agent.tools.builtin_data_analysis_tools import create_chart_from_plotly_config
import json
# Define Plotly configuration
plotly_config = {
"data": [
{
"type": "bar",
"x": ["Q1", "Q2", "Q3", "Q4"],
"y": [100, 150, 120, 180],
"name": "Revenue"
}
],
"layout": {
"title": "Quarterly Revenue 2024",
"xaxis": {"title": "Quarter"},
"yaxis": {"title": "Revenue ($K)"}
}
}
# Create chart
result = await create_chart_from_plotly_config(
config_content=json.dumps(plotly_config),
config_format="json",
output_filename="revenue_chart",
output_format="png",
tool_context=context
)
if result.status == "success":
# Chart is saved as artifact and can be displayed
chart_data = result.data_objects[0]
print(f"Chart created: {chart_data.name}")
YAML Configuration Example
yaml_config = """
data:
- type: scatter
x: [1, 2, 3, 4, 5]
y: [10, 15, 13, 17, 20]
mode: lines+markers
name: Sales Trend
layout:
title: Sales Over Time
xaxis:
title: Month
yaxis:
title: Sales
"""
result = await create_chart_from_plotly_config(
config_content=yaml_config,
config_format="yaml",
output_filename="sales_trend.png",
output_format="png",
tool_context=context
)
Supported Chart Types
Plotly supports a wide variety of chart types:
- Bar Charts: Horizontal or vertical bars
- Line Charts: Connected data points
- Scatter Plots: Individual data points
- Pie Charts: Proportional circular sectors
- Area Charts: Filled regions under lines
- Box Plots: Distribution summaries
- Violin Plots: Distribution shapes
- Histograms: Frequency distributions
- Heatmaps: 2D data matrices
- 3D Plots: Surface, scatter, and line plots in 3D
- Contour Plots: Topographic-style visualizations
- Candlestick: Financial OHLC data
- Funnel Charts: Stage-based visualizations
- Sunburst: Hierarchical data
Complete Example: Sales Dashboard
import json
# Multi-series bar chart with custom styling
config = {
"data": [
{
"type": "bar",
"name": "2023",
"x": ["Jan", "Feb", "Mar", "Apr"],
"y": [120, 135, 145, 160],
"marker": {"color": "#3b82f6"}
},
{
"type": "bar",
"name": "2024",
"x": ["Jan", "Feb", "Mar", "Apr"],
"y": [140, 155, 165, 185],
"marker": {"color": "#10b981"}
}
],
"layout": {
"title": {
"text": "Sales Comparison: 2023 vs 2024",
"font": {"size": 20}
},
"xaxis": {"title": "Month"},
"yaxis": {"title": "Sales ($K)"},
"barmode": "group",
"template": "plotly_white",
"showlegend": True,
"width": 800,
"height": 500
}
}
result = await create_chart_from_plotly_config(
config_content=json.dumps(config, indent=2),
config_format="json",
output_filename="sales_comparison.png",
output_format="png",
tool_context=context
)
PNG - Raster image format, good for web and presentations. Best for most use cases.
SVG - Vector format, scalable without quality loss. Ideal for publications and print.
PDF - Vector format in PDF container. Perfect for reports and documents.
JPEG - Compressed raster format. Use when file size is critical.
WebP - Modern web format with better compression than PNG/JPEG.
Configuration
The chart creation tool requires the plotly and kaleido libraries:
builtin_tools:
- name: create_chart_from_plotly_config
Required Dependencies
pip install plotly kaleido
Kaleido is required for image export. If not installed, you’ll receive an error message.
DataObject in the result:
{
"status": "success",
"message": "Chart image created successfully.",
"data": {
"source_format": "json",
"output_format": "png"
},
"data_objects": [
{
"name": "revenue_chart.png",
"mime_type": "image/png",
"disposition": "artifact",
"description": "Chart generated from Plotly config (json)",
"metadata": {
"source_format": "json",
"output_format": "png",
"generation_tool": "create_chart_from_plotly_config",
"generation_timestamp": "2024-03-01T10:30:00Z"
}
}
]
}
Advanced Examples
Heatmap with Custom Colorscale
config = {
"data": [
{
"type": "heatmap",
"z": [
[1, 20, 30, 50, 1],
[20, 1, 60, 80, 30],
[30, 60, 1, -10, 20]
],
"x": ["Mon", "Tue", "Wed", "Thu", "Fri"],
"y": ["Morning", "Afternoon", "Evening"],
"colorscale": "Viridis"
}
],
"layout": {
"title": "Activity Heatmap"
}
}
3D Surface Plot
import numpy as np
# Generate data
x = np.linspace(-5, 5, 50)
y = np.linspace(-5, 5, 50)
X, Y = np.meshgrid(x, y)
Z = np.sin(np.sqrt(X**2 + Y**2))
config = {
"data": [
{
"type": "surface",
"z": Z.tolist(),
"colorscale": "Viridis"
}
],
"layout": {
"title": "3D Surface Plot",
"scene": {
"xaxis": {"title": "X"},
"yaxis": {"title": "Y"},
"zaxis": {"title": "Z"}
}
}
}
Error Handling
result = await create_chart_from_plotly_config(
config_content=invalid_json,
config_format="json",
output_filename="chart.png",
tool_context=context
)
if result.status == "error":
if "parse" in result.message.lower():
print("Invalid JSON/YAML configuration")
elif "plotly" in result.message.lower():
print("Invalid Plotly configuration")
elif "kaleido" in result.message.lower():
print("Kaleido library not installed")
Best Practices
Configuration Tips
- Always validate your Plotly config before passing to the tool
- Use descriptive titles and axis labels
- Choose appropriate chart types for your data
- Consider output format based on use case (PNG for web, SVG for print)
- Set appropriate width/height in layout for better rendering
See Also
