Skip to main content

​
Playground

The Bindu Playground is an interactive environment where you can experiment with agents, test protocols, and explore multi-agent systems without writing code.
The playground is coming soon! In the meantime, you can test agents locally using the quickstart guide.

​
Features

Interactive Testing

Test agents with real-time feedback

Protocol Explorer

Explore A2A and AP2 protocol features

Multi-Agent Scenarios

Simulate agent collaboration

Visual Debugging

See message flows and state changes

​
What You Can Do

​
Test Agent Responses

Send messages to agents and see how they respond: 
{
  "message": "Explain quantum computing in simple terms",
  "agent": "research-agent",
  "model": "gpt-4o"
}

​
Visualize Agent Communication

See how agents communicate using the Bindu protocol:

​
Experiment with Tools

Test different tools and see how agents use them:
  • Web search
  • File operations
  • API calls
  • Database queries
  • Custom functions

​
Monitor Performance

Track agent metrics in real-time:
  • Response time
  • Token usage
  • Success rate
  • Error frequency

​
Coming Soon

Visual interface to create and configure agents without code
Design multi-agent workflows with drag-and-drop
Automated testing for agent behavior and responses
Share playground sessions with team members
Pre-built agent templates for common use cases

​
Local Testing

While the playground is in development, you can test agents locally:

​
Quick Test Script

import requests
import uuid

def test_agent(message: str, agent_url: str = "http://localhost:8030"):
    payload = {
        "jsonrpc": "2.0",
        "method": "message/send",
        "params": {
            "message": {
                "role": "user",
                "parts": [{"kind": "text", "text": message}],
                "kind": "message",
                "messageId": str(uuid.uuid4()),
                "contextId": str(uuid.uuid4()),
                "taskId": str(uuid.uuid4())
            },
            "configuration": {
                "acceptedOutputModes": ["application/json"]
            }
        },
        "id": str(uuid.uuid4())
    }
    
    response = requests.post(agent_url, json=payload)
    return response.json()

# Test your agent
result = test_agent("Hello, how are you?")
print(result)

​
Interactive CLI

# Install Bindu CLI (coming soon)
pip install bindu-cli

# Start interactive session
bindu playground

# Test agent
> send "What is the weather today?"

# List available agents
> agents list

# Switch agent
> use research-agent

# View agent card
> agent info

​
Use Cases

​
Learning

  • Understand how agents process messages
  • Explore protocol features
  • Test different models and configurations

​
Development

  • Prototype agent behavior
  • Test tool integrations
  • Debug agent interactions

​
Demonstration

  • Show agent capabilities to stakeholders
  • Create interactive demos
  • Present multi-agent workflows

​
Best Practices

Start with simple single-agent tests before moving to complex multi-agent scenarios.
  1. Start Simple: Test basic functionality first
  2. Iterate Quickly: Make small changes and test immediately
  3. Save Scenarios: Keep successful test cases for regression testing
  4. Monitor Costs: Track token usage when testing with paid models
  5. Document Findings: Note interesting behaviors and edge cases

​
Next Steps

Quickstart

Build your first agent

Examples

See example implementations

API Reference

Explore the API

Monitoring

Learn about monitoring
⌘I