py-agent-client 0.0.2.dev1

Placeholder for the official Python SDK for the Intelligent Agents Platform (IAP). Coming soon!

🚀 py-agent-client

Build powerful MCP client applications with the Intelligent Agents Platform

🎯 What is py-agent-client?

py-agent-client is the flagship client library for the Intelligent Agents Platform (IAP). It enables developers to build applications that interact seamlessly with Model Context Protocol (MCP) servers, making AI agent development intuitive and powerful.

✨ Key Features:

• 🔌 Universal MCP Support - Connect to any MCP-compliant server
• ⚡ High Performance - Optimized for speed and reliability
• 🛠️ Rich Tooling - Comprehensive tool discovery and execution
• 🐍 Pythonic Design - Clean, intuitive API
• 🔒 Enterprise Ready - Built for production environments

🚀 Quick Start

Installation

pip install py-agent-client

Your First MCP Client

from py_agent_client import MCPClient

async def main():
    # Connect to an MCP server
    client = MCPClient()
    await client.connect("ws://localhost:8080")
    
    # Discover available tools
    tools = await client.list_tools()
    print(f"Found {len(tools)} tools available")
    
    # Execute a tool
    result = await client.call_tool("file_reader", {
        "path": "document.txt"
    })
    print(f"Result: {result}")

import asyncio
asyncio.run(main())

🏗️ Intelligent Agents Platform Ecosystem

py-agent-client is part of the comprehensive IAP ecosystem:

🏗️ py-agent-core       - Foundation & shared utilities
├── py-agent-client    - ✅ Client applications (You are here)
├── py-agent-server    - 🚧 MCP server development
├── py-agent-tool      - 🚧 Development & debugging tools
└── py-agent-resources - 🚧 Templates & examples

🔧 Features & Examples

Tool Discovery

# List all available tools
tools = await client.list_tools()

# Get detailed tool information
tool_info = await client.get_tool_info("web_scraper")
print(f"Parameters: {tool_info.parameters}")

Tool Execution

# Execute tools with parameters
result = await client.call_tool("calculator", {
    "operation": "multiply",
    "a": 15,
    "b": 7
})

# Handle errors gracefully
try:
    data = await client.call_tool("api_caller", {
        "url": "https://api.example.com/data"
    })
except MCPToolError as e:
    print(f"Tool execution failed: {e}")

Advanced Connection Management

from py_agent_client import ConnectionConfig

# Custom configuration
config = ConnectionConfig(
    url="wss://secure-server.com:8443",
    timeout=30,
    retry_attempts=5,
    headers={"Authorization": "Bearer token123"}
)

client = MCPClient(config)
await client.connect()

Real-time Events

# Listen for server events
@client.on("tool_added")
async def on_new_tool(tool_info):
    print(f"New tool available: {tool_info.name}")

@client.on("connection_lost")
async def on_disconnect():
    await client.reconnect()

📊 Monitoring & Performance

Built-in Metrics

# Get performance statistics
stats = client.get_stats()
print(f"Tools called: {stats.tools_called}")
print(f"Average response time: {stats.avg_response_time}ms")
print(f"Success rate: {stats.success_rate}%")

Custom Logging

import logging

# Enable debug logging
logging.getLogger("py_agent_client").setLevel(logging.DEBUG)

🔧 Configuration

Environment Variables

export IAP_SERVER_URL="ws://localhost:8080"
export IAP_TIMEOUT=30
export IAP_API_KEY="your_api_key"

Configuration File

# config.py
IAP_CONFIG = {
    "server_url": "ws://localhost:8080",
    "timeout": 30,
    "retry_attempts": 3,
    "tools": {
        "enabled": ["file_ops", "web_scraper", "calculator"]
    }
}

🚧 Ecosystem Status

Package	Status	Description
py-agent-client	✅ Available Now	Full-featured MCP client
py-agent-core	🚧 In Development	Foundation & shared utilities
py-agent-server	🚧 In Development	Build MCP servers
py-agent-tool	🚧 In Development	Development & debugging tools
py-agent-resources	🚧 In Development	Templates & examples

📚 Resources

• 📖 Documentation
• 🎯 Examples
• 🐛 Issues
• 💬 Discussions

🤝 Contributing

Contributions are welcome! Please visit our GitHub repository for contribution guidelines.

Development Setup

git clone https://github.com/fmonfasani/intelligent-agents-platform
cd intelligent-agents-platform
pip install -e .[dev,test]
pytest tests/

📄 License

Licensed under the MIT License - see LICENSE for details.

---

🚀 Start building intelligent agents today!

Built with ❤️ by Federico Monfasani

Part of the Intelligent Agents Platform