Quick Start¶

This guide will get you up and running with MCPOmni Connect in under 5 minutes.

Step 1: Basic Configuration¶

Create the two required configuration files:

Create .env file¶

# Create .env file with your LLM API key
echo "LLM_API_KEY=your_api_key_here" > .env

Create servers_config.json¶

cat > servers_config.json << 'EOF'
{
    "AgentConfig": {
        "tool_call_timeout": 30,
        "max_steps": 15,
        "request_limit": 1000,
        "total_tokens_limit": 100000
    },
    "LLM": {
        "provider": "openai",
        "model": "gpt-4o-mini",
        "temperature": 0.5,
        "max_tokens": 5000,
        "top_p": 0.7
    },
    "mcpServers": {}
}
EOF

Step 2: Start MCPOmni Connect¶

mcpomni_connect

You should see the MCPOmni Connect CLI start up:

🚀 MCPOmni Connect - Universal Gateway to MCP Servers
Connected to 0 MCP servers
Mode: CHAT (type /mode:auto for autonomous mode)

> 

Step 3: Test Basic Functionality¶

Try these commands to verify everything is working:

Check Available Commands¶

/help

Test LLM Connection¶

Hello! Can you tell me about yourself?

The AI should respond, confirming your LLM configuration is working.

Step 4: Add Your First MCP Server¶

Let's add a simple MCP server to demonstrate connectivity:

Option A: File System Server (Local)¶

Edit your servers_config.json to add a file system server:

{
    "AgentConfig": {
        "tool_call_timeout": 30,
        "max_steps": 15,
        "request_limit": 1000,
        "total_tokens_limit": 100000
    },
    "LLM": {
        "provider": "openai",
        "model": "gpt-4o-mini",
        "temperature": 0.5,
        "max_tokens": 5000,
        "top_p": 0.7
    },
    "mcpServers": {
        "filesystem": {
            "transport_type": "stdio",
            "command": "uvx",
            "args": ["mcp-server-filesystem", "/tmp"]
        }
    }
}

Option B: Remote HTTP Server¶

{
    "mcpServers": {
        "remote-server": {
            "transport_type": "streamable_http",
            "url": "http://your-server.com:8080/mcp",
            "headers": {
                "Authorization": "Bearer your-token"
            },
            "timeout": 60
        }
    }
}

Restart and Test¶

# Restart MCPOmni Connect
mcpomni_connect

Now check available tools:

/tools

You should see tools from your connected MCP server!

Step 5: Try Different Operation Modes¶

Chat Mode (Default)¶

Can you list the files in the current directory?

Autonomous Mode¶

/mode:auto
Can you analyze the files in the current directory and create a summary?

Switch Back to Chat Mode¶

/mode:chat

Common First Tasks¶

Explore Available Capabilities¶

/tools      # List all available tools
/prompts    # Show available prompts
/resources  # Display available resources

Memory Management¶

/memory     # Toggle Redis memory persistence

Debug Mode¶

/debug      # Enable detailed logging for troubleshooting

Next Steps¶

Now that you have MCPOmni Connect running:

1. Configure additional LLM providers - Try different AI models
2. Add more MCP servers - Connect to databases, APIs, and tools
3. Explore advanced features - Learn about ReAct agents and orchestration
4. Set up authentication - Configure OAuth and secure connections

Troubleshooting Quick Start¶

Congratulations! 🎉 You now have MCPOmni Connect running. Ready to explore more advanced features?

Next: Configuration Guide →