Playground Page

​

Overview

​

Key Features

• Real-time API Testing: Execute Swarms API calls directly in the browser
• Multi-language Code Generation: Generate code examples in Python, Rust, Go, and TypeScript
• Interactive Configuration: Visual interface for setting up agent parameters
• Live Output: See API responses immediately in the output terminal
• Code Export: Copy generated code for use in your applications

​

Interface Overview

​

Language Selection

• Python: Default language with requests library implementation
• Rust: Native Rust HTTP client implementation
• Go: Standard Go HTTP package implementation
• TypeScript: Node.js/browser-compatible implementation

​

Agent Modes

​

Single Agent Mode

• Prototype testing
• Parameter optimization
• Simple task automation
• API familiarization

​

Multi-Agent Mode

• Complex workflow automation
• Collaborative AI systems
• Distributed task processing
• Advanced orchestration scenarios

​

Configuration Parameters

​

Basic Agent Settings

​

Agent Name

​

Model Name

​

Description

​

System Prompt

• Be clear and specific
• Include role definition
• Specify output format if needed
• Add relevant constraints

​

Advanced Parameters

​

Temperature

• Low (0.0-0.3): More deterministic, consistent responses
• Medium (0.4-0.7): Balanced creativity and consistency
• High (0.8-2.0): More creative and varied responses

​

Max Tokens

• Higher values allow longer responses
• Impacts API costs
• Model-dependent limits apply

​

Role

​

Max Loops

• 1: Single response
• >1: Allows iterative problem solving

​

MCP URL (Optional)

​

Task Definition

​

Task

• Be specific and clear
• Include all necessary context
• Specify desired output format
• Provide examples when helpful

​

Using the Playground

​

Step-by-Step Guide

1. Select Mode: Choose between Single Agent or Multi-Agent
2. Choose Language: Select your preferred programming language
3. Configure Agent: Fill in the required parameters
4. Define Task: Enter your specific task or query
5. Run Agent: Click the “Run Agent” button
6. Review Output: Check the Output Terminal for results
7. Copy Code: Use the generated code in your applications

​

Testing Strategies

​

Parameter Experimentation

• Temperature Testing: Try different temperature values to find optimal creativity levels
• Prompt Engineering: Iterate on system prompts to improve responses
• Token Optimization: Adjust max_tokens based on expected response length

​

Workflow Development

• Start Simple: Begin with basic tasks and gradually increase complexity
• Iterative Refinement: Use playground results to refine your approach
• Documentation: Keep notes on successful configurations

​

Output Interpretation

​

Output Terminal

• Agent Responses: Direct output from the AI agent
• Error Messages: API errors or configuration issues
• Execution Status: Success/failure indicators
• Response Metadata: Token usage, timing information

​

Code Preview

• Complete Implementation: Ready-to-use code in your selected language
• API Configuration: Proper headers and authentication setup
• Request Structure: Correctly formatted payload
• Response Handling: Basic error handling and output processing

​

Code Examples by Language

​

Python Implementation

import requests

url = "https://swarms-api-285321057562.us-east1.run.app/v1/agent/completions"
headers = {
    "Content-Type": "application/json",
    "x-api-key": "your-api-key-here"
}

payload = {
    "agent_config": {
        "agent_name": "example_agent",
        "description": "Example agent for demonstration",
        "system_prompt": "You are a helpful assistant.",
        "model_name": "gpt-4o-mini",
        "auto_generate_prompt": false,
        "max_tokens": 8192,
        "temperature": 0.5,
        "role": "worker",
        "max_loops": 1,
        "tools_list_dictionary": null,
        "mcp_url": null
    },
    "task": "Explain quantum computing in simple terms"
}

response = requests.post(url, json=payload, headers=headers)
print(response.json())

​

Key Code Components

​

API Endpoint

• URL: https://swarms-api-285321057562.us-east1.run.app/v1/agent/completions
• Method: POST
• Authentication: API key in x-api-key header

​

Request Structure

• Headers: Content-Type and API key
• Payload: Agent configuration and task
• Response: JSON with agent output and metadata

​

Best Practices

​

Security

• API Key Management: Never expose API keys in client-side code
• Environment Variables: Store sensitive credentials securely
• Rate Limiting: Respect API rate limits in production

​

Performance Optimization

• Parameter Tuning: Optimize temperature and max_tokens for your use case
• Prompt Engineering: Craft efficient system prompts
• Caching: Implement response caching for repeated queries

​

Development Workflow

• Prototype in Playground: Test configurations before implementation
• Document Successful Configs: Save working parameter combinations
• Iterate and Improve: Use playground for continuous optimization

​

Troubleshooting

​

Common Issues

​

No Output in Terminal

• Check API Key: Ensure valid API key is configured
• Verify Parameters: All required fields must be filled
• Network Issues: Check internet connection

​

Unexpected Responses

• Review System Prompt: Ensure clear instructions
• Adjust Temperature: Try different creativity levels
• Check Task Definition: Verify task clarity and specificity

​

Code Generation Issues

• Language Selection: Ensure correct language is selected
• Copy Functionality: Use the “Copy Code” button for accurate copying
• Syntax Validation: Test generated code in your development environment

​

Integration Guide

​

From Playground to Production

1. Copy Generated Code: Use the Code Preview section
2. Add Error Handling: Implement robust error handling
3. Configure Environment: Set up proper API key management
4. Test Thoroughly: Validate in your target environment
5. Monitor Performance: Track API usage and response quality