Skip to main content

​
What is the Playground?

The Playground is an enhanced testing environment in the Advanced Editor where you can interact with your agent in real-time. It allows you to simulate conversations, test different scenarios, and validate your agent’s behavior before deploying to production, all while leveraging the improved code editor features.
Testing in the Playground saves time and money by catching issues before they affect real customer calls, with enhanced debugging capabilities through the improved editor interface.

​
Enhanced Editor Integration

The Playground now benefits from the Advanced Editor’s improved interface, including:
  • View Controls: Use outline, indented view, and minimap for better code navigation
  • Diff Mode: Compare changes with inline/split view options when working with versions
  • Focus Mode: Minimize distractions while testing
  • Version Integration: See how changes from different versions affect playground behavior

​
Accessing the Playground

The Playground is located in the right panel of the Advanced Editor. Click the “Playground” tab to access it.
1

Navigate to Advanced Editor

Open your agent in the Advanced Editor.
2

Open Playground

Click the “Playground” tab in the right panel.
3

Configure View

Use the editor view controls (outline, minimap, focus mode) to optimize your workspace.

​
Call Types

You can simulate two types of calls with enhanced testing capabilities:
TypeDescriptionUse CaseEditor Features
📞 InboundCustomer calling your agentTest welcome flow, support scenariosUse diff mode to compare different greeting versions
📤 OutboundAgent calling a customerTest introductions, sales scriptsFocus mode helps concentrate on script testing
Switch between call types to test both greetings. Use the editor’s version history to compare how different versions handle each scenario.

​
Enhanced Testing Interface

​
Code Editor Improvements

The Playground now features:
  • Outline View: Navigate through your agent’s structure quickly
  • Indented View: Better visualization of nested logic
  • Minimap: Overview of your entire agent configuration
  • Focus Mode: Hide distractions during testing

​
Sending Messages

​
Text Input

Type messages in the input field and press Enter or click Send. The agent will respond as it would in a real call, with enhanced debugging information available through the improved editor interface.

​
Voice Input

Click the microphone icon to speak your message. The agent will:
  1. Transcribe your speech
  2. Process the message
  3. Respond with synthesized voice
Voice testing now integrates with the version system - changes to voice settings are immediately reflected in testing.

​
Enhanced Conversation View

The Playground displays messages chronologically with improved visualization:
ElementDescriptionEditor Enhancement
👤 User MessageYour inputs (text or transcribed voice)Syntax highlighting for better readability
🤖 Agent MessageAgent responsesDiff highlighting when comparing versions
🔧 Tool CallWhen agent triggers a webhook or actionCollapsible view with detailed parameters
✅ Tool ResultResponse from webhook or actionJSON formatting with syntax highlighting

​
Audio Playback with Version Tracking

For each agent response, you can:
  • ▶️ Play the audio response
  • ⏸️ Pause playback
  • 🔁 Replay messages
  • 📊 Compare audio quality across versions

​
Context Variables with Version Control

Test how your agent handles different input parameters with enhanced management:
1

Open Context Variables

Click the “Context Variables” panel in the Playground.
2

Configure Variables

Set test values with version-aware validation.
3

Version Comparison

Use diff mode to see how variable handling changed between versions.
4

Start Testing

Begin testing with the configured context.

​
Enhanced Context Management

{
  "customer_name": "John Smith",
  "order_number": "ORD-12345", 
  "is_premium_customer": true,
  "account_balance": 1250.50,
  "_version_context": {
    "version": "v1.2.3",
    "changes": ["Updated premium logic", "New balance validation"]
  }
}

​
Reliability Test with Version Integration

Run multiple iterations with enhanced tracking:
1

Configure Scenario

Set up your test conversation and context variables.
2

Select Version

Choose which version to test or compare multiple versions.
3

Set Iterations

Choose how many times to run the test (e.g., 10, 50, 100).
4

Run Enhanced Test

Click “Run Reliability Test” with version tracking enabled.
5

Analyze Results

Review consistency with version-specific insights.
Reliability tests now track version-specific performance metrics. Results may vary between versions due to model or configuration changes.

​
Enhanced Testing Features

​
Advanced Tool Call Testing

When your agent triggers a webhook:
  1. Enhanced Visualization: Tool calls appear with syntax highlighting
  2. Parameter Inspection: Detailed parameter view with diff comparison
  3. Response Analysis: JSON formatting with collapsible sections
  4. Version Comparison: See how tool behavior changed between versions

​
Evaluation Testing with Diff Support

After a conversation, run evaluations with enhanced analysis:
  1. End the conversation naturally
  2. Click “Run Evaluations”
  3. New: Compare evaluation results across versions
  4. Review which evaluations pass or fail with detailed diff highlighting

​
Knowledge Base Testing with Version Tracking

Verify your agent’s knowledge retrieval with version awareness:
  1. Ask questions that require KB lookup
  2. Observe retrieval with enhanced debugging information
  3. New: Compare knowledge retrieval across different versions
  4. Check response accuracy with diff highlighting

​
Enhanced Best Practices

Use the version system to test changes systematically. Compare behavior across versions using diff mode.
Enable focus mode when doing deep testing sessions to minimize distractions and concentrate on agent behavior.
Multiple team members can test simultaneously with real-time sync showing current version state.
Use the improved editor interface to identify and test more complex edge cases with better visualization.
Use inline/split diff modes to understand exactly how changes affect agent behavior during testing.

​
Advanced Testing Scenarios

​
Version Regression Testing

Test that new versions don’t break existing functionality:
  • Load previous version in split diff view
  • Run same test scenarios
  • Compare responses side-by-side

​
Collaborative Testing Sessions

Multiple team members testing simultaneously:
  • Real-time version sync across all playground instances
  • Shared context variables and test scenarios
  • Live collaboration indicators

​
Performance Testing with Version Metrics

Track performance changes across versions:
  • Response time comparisons
  • Memory usage tracking (in debug mode)
  • Token consumption analysis

​
Enhanced Troubleshooting

IssueEnhanced SolutionEditor Features
Agent not respondingCheck LLM config with syntax highlightingUse outline view to navigate to model settings
Wrong greetingVerify call type with diff comparisonCompare greeting logic across versions
Missing personalizationContext variables with version validationUse diff mode to see variable handling changes
Webhook issuesEnhanced tool debugging with JSON formattingCollapsible parameter inspection
Version conflictsMerge conflict resolution in playgroundInline diff shows conflicting changes

​
Integration with Version System

The Playground now integrates deeply with the version control system:
  • Real-time Sync: Testing always reflects the current version state
  • Change Detection: Visual indicators show when testing results might differ due to recent changes
  • Conflict Resolution: Test behavior during merge conflicts to ensure smooth transitions

​
Next Steps

Copilot Integration

Get AI assistance with enhanced prompt analysis from playground results

Version History

Track how changes affect playground behavior over time