Examples

Comprehensive examples demonstrating haive-mcp capabilities with 1900+ MCP servers.

Example Scripts

Our examples directory contains working scripts that demonstrate various MCP integration patterns:

Basic Examples

  • basic_mcp_agent.py - Simple agent with MCP tools

  • basic_download.py - Basic file download operations

  • simple_discovery_demo.py - Discovering MCP servers

  • simple_test_discovery.py - Testing discovery functionality

Advanced Integration

  • aug_llm_mcp_integration.py - Integration with AugLLM engine

  • complete_mcp_integration.py - Full MCP integration example

  • dynamic_mcp_agent_system.py - Dynamic agent creation

  • typed_mcp_aug_llm_example.py - Type-safe MCP integration

Automation & Discovery

  • automated_discovery_agent.py - Automatic server discovery from 1900+ servers

  • comprehensive_mcp_discovery.py - Comprehensive discovery patterns

  • dynamic_activation_mcp_example.py - Dynamic tool activation

  • procedural_mcp_addition.py - Adding MCP servers procedurally

Production Examples

  • production_mcp_harvester.py - Production-ready MCP harvesting

  • background_mcp_processor.py - Background processing with MCP

  • mcp_integration_pipeline.py - Complete integration pipeline

  • documentation_processing_pipeline.py - Processing documentation with MCP

Specialized Use Cases

  • ai_enhanced_coding.py - AI-powered coding assistance

  • dataflow_mcp_example.py - Dataflow integration

  • batch_operations.py - Batch processing with MCP

  • custom_installer.py - Custom server installation

Quick Example

Here’s a simple example to get started:

from haive.mcp.agents.enhanced_mcp_agent import EnhancedMCPAgent
from haive.core.engine.aug_llm import AugLLMConfig
import asyncio

async def main():
    # Create agent with automatic MCP discovery
    agent = EnhancedMCPAgent(
        name="example_agent",
        engine=AugLLMConfig(temperature=0.7),
        mcp_categories=["core", "ai_enhanced"],
        auto_install=True
    )

    # Initialize (auto-installs servers)
    await agent.initialize_mcp()

    # Use the agent
    result = await agent.arun("Analyze Python files and create a summary")
    print(result)

asyncio.run(main())

Running Examples

To run any example:

# Navigate to examples directory
cd docs/source/examples

# Run an example
poetry run python basic_mcp_agent.py

# Or run with the shell script
./run_mcp_agent.sh

Configuration

Some examples use the mcp_servers_config.json file for configuration. This file defines server categories and installation preferences.

Next Steps

  • Review Tutorials for step-by-step guides

  • Check Guides for architectural patterns

  • Explore the API Reference for detailed documentation