Tutorial: Getting Started with Panel Live Server¶

In this tutorial you'll learn how to use Panel Live Server to create, view, and share interactive Python visualizations. You'll explore both standalone usage and AI-assisted workflows. By the end, you'll have created multiple visualizations and understand how to integrate Panel Live Server into your development workflow.

What You'll Learn¶

• Install and start Panel Live Server
• Create visualizations using the web interface (standalone mode)
• Create visualizations using AI assistants (MCP tool mode)
• View, browse, and manage your visualizations
• Understand execution methods (Jupyter vs Panel)
• Create visualizations programmatically using the REST API

What You'll Need¶

• Python 3.12 or later
• Panel Live Server installed (pip install panel-live-server)
• Basic familiarity with Python and data visualization
• A web browser
• (Optional) An AI assistant configured with Panel Live Server for Part 2

Understanding Panel Live Server¶

Panel Live Server has two components:

1. Panel Server (pls serve): A local web server that runs visualizations and provides a browser interface
2. MCP Server (pls mcp): An MCP server that lets AI assistants create visualizations via the show tool

You can use Panel Live Server in two ways:

• Standalone: Start the server manually and create visualizations via the web interface or REST API
• With AI: Start the MCP server — the Panel server starts automatically, and AI assistants use it via the show tool

Part 1: Using Panel Live Server Standalone¶

Step 1: Install Panel Live Server¶

pip install panel-live-server

Step 2: Start the Server¶

Open your terminal and start Panel Live Server:

pls serve

You should see output like this:

Starting Panel Live Server...
Panel Live Server running at:

  Add:   http://localhost:5077/add
  Feed:  http://localhost:5077/feed
  Admin: http://localhost:5077/admin

Your server is now running. Keep this terminal open while you work through the tutorial.

pls serve --port 9999
pls serve --host 0.0.0.0 --port 8080

Step 3: Create Your First Visualization¶

Open your web browser and navigate to http://localhost:5077/add. You'll see a form with a code editor.

Enter the following Python code:

import pandas as pd
import hvplot.pandas

df = pd.DataFrame({
    'Product': ['A', 'B', 'C', 'D'],
    'Sales': [120, 95, 180, 150]
})

df.hvplot.bar(x='Product', y='Sales', title='Sales by Product')

Fill in the form fields:

• Name: "Product Sales Chart"
• Description: "An interactive bar chart showing sales by product"
• Execution Method: jupyter (default)

Click Submit. You'll see a success message with a link to your visualization.

Step 4: View Your Visualization¶

Click the link in the success message. You'll be taken to a URL like http://localhost:5077/view?id=abc123 where your chart is displayed.

The bar chart is interactive — hover over the bars to see tooltips.

Step 5: Browse Your Visualizations¶

Navigate to http://localhost:5077/feed to see a live-updating list of your recent visualizations, including names, descriptions, and inline previews.

Step 6: Manage Your Collection¶

Visit http://localhost:5077/admin for a table view of all your snippets. From here you can search, filter, and delete visualizations.

Step 7: Create Visualizations Programmatically¶

You can also create visualizations via the REST API. Create script.py:

import requests

response = requests.post(
    "http://localhost:5077/api/snippet",
    headers={"Content-Type": "application/json"},
    json={
        "code": "a = 'Hello, Panel Live Server!'\na",
        "name": "Hello World",
        "method": "jupyter"
    }
)

print(f"Status: {response.status_code}")
print(f"Response: {response.json()}")

Run it:

python script.py

The response contains the URL of your new visualization.

Part 2: Using Panel Live Server with AI Assistants¶

Now let's use the MCP server to create visualizations through AI assistants with natural language.

Prerequisites¶

An AI assistant (Claude, GitHub Copilot, etc.) configured to use the Panel Live Server MCP server. See the configure server guide for setup instructions.

Step 1: Start the MCP Server¶

In your IDE, start the Panel Live Server MCP server. The Panel server starts automatically.

Step 2: Create Your First AI-Assisted Visualization¶

Open your AI assistant and ask:

My dataset is penguins.csv. What is the distribution of the 'species' column? Use the show tool.

Your AI assistant will use the show tool and respond with:

Visualization created successfully!
View at: http://localhost:5077/view?id=...

Click the URL to see an interactive bar chart showing penguin species counts.

Step 3: Explore Relationships with Scatter Plots¶

Ask your AI:

Show me a scatter plot of 'flipper_length_mm' vs 'body_mass_g'

The AI creates a scatter plot with interactive tooltips, zoom, and pan.

Step 4: Build Interactive Dashboards¶

For advanced use cases:

Create an interactive dashboard for the penguins dataset with dropdown filters for species and island.

The AI will produce a Panel app with reactive widgets — dropdowns that filter the chart in real time.

Step 5: Refine Your Visualizations¶

If the result isn't what you expected, continue the conversation:

• "Color the points by species"
• "Add a trend line to the scatter plot"
• "Show only penguins with body mass greater than 4000g"
• "Display these two charts side by side"

The AI iterates on your work, creating new visualizations that build on previous ones.

Understanding Execution Methods¶

Jupyter Method (Default)¶

Executes code like a Jupyter notebook cell — the last expression is automatically displayed:

import pandas as pd
df = pd.DataFrame({'x': [1, 2, 3], 'y': [4, 5, 6]})
df  # This is displayed

Use this for data exploration, quick charts, and simple visualizations.

Panel Method¶

For Panel dashboard applications with explicit .servable() calls:

import panel as pn

pn.extension()

pn.Column(
    pn.pane.Markdown("# My Dashboard"),
    pn.widgets.Button(name="Click me")
).servable()

Use this for complex, interactive applications with multiple components.

Understanding Storage¶

All visualizations are stored in a local SQLite database:

~/.panel-live-server/snippets/snippets.db

The database stores your Python code, execution results, metadata (names, descriptions, timestamps), and detected packages.

Checking Server Status¶

pls status

This queries the health endpoint and reports whether the server is running.

Troubleshooting¶

ModuleNotFoundError¶

Install the missing package — no server restart required:

pip install package-name

Server Not Available (MCP mode)¶

Verify the MCP server is running and check its startup logs for "Panel server started successfully".

Visualization Shows an Error¶

Ask your AI to fix it based on the error message shown, or start with a simpler visualization to confirm the system is working.

What You've Learned¶

• Start Panel Live Server standalone or via MCP
• Create visualizations using the web interface, REST API, and AI assistants
• View, browse, and manage your visualizations
• Use Jupyter and Panel execution methods
• Build interactive dashboards with natural language

Next Steps¶

• Architecture — understand how the server, MCP, and browser work together
• Configure the server — customize ports, host, database path, and restart limits
• Examples — quick code snippets to copy and adapt