Consulting Agents

🚀 Guide to Using the Consulting Agent Model Console Server

🚀 Quick Start

This guide provides detailed steps on how to install, configure, and use the Consulting Agent Model Console (MCP) Server.

📦 Installation

To use the Consulting Agent Model Console (MCP) Server, follow these installation steps:

1. Clone the code repository:

   git clone https://github.com/your-username/consulting-agents-mcp.git
   cd consulting-agents-mcp
   

2. Install dependencies: Open a terminal and run the following command to install the required Python packages:

   pip install -r requirements.txt
   

🛠️ Configuration

Environment Variables

Before running the MCP Server, set the following environment variables:

• MCP_TRANSPORT: Transport protocol (default value is stdio, optional values are http or sse).
• HOST: Host address when using HTTP/SSE transport (default value is 127.0.0.1).
• PORT: Port number when using HTTP/SSE transport (default value is 5000).

💻 Usage Examples

Consulting Agent Functions

Darren (Darren Agent)

Darren is an advanced consulting agent focusing on complex model inference tasks.

Basic Usage

{
  "agent": "Darren",
  "consultation_context": "Describe your consulting request...",
  "source_code": "Provide relevant code..."
}

Advanced Usage

curl -X POST http://localhost:5000/consult \
-H "Content-Type: application/json" \
-d '{"agent":"Darren","consultation_context":"How can I optimize the inference speed of my model?","source_code":"import torch\nmodel = torch.load('my_model.pth')"}'

Sonny (Sonny Agent)

Sonny is an agent focusing on code analysis and improvement suggestions.

Basic Usage

{
  "agent": "Sonny",
  "consultation_context": "Describe your consulting request...",
  "source_code": "Provide relevant code..."
}

Advanced Usage

curl -X POST http://localhost:5000/consult \
-H "Content-Type: application/json" \
-d '{"agent":"Sonny","consultation_context":"My Python code has low running efficiency.","source_code":"def slow_function():\n    for i in range(1000000):\n        print(i)"}'

Sergey (Sergey Agent)

Sergey is a search engine expert, good at solving problems through web searches.

Basic Usage

{
  "agent": "Sergey",
  "consultation_context": "Describe your consulting request...",
  "search_query": "Specify search keywords..."
}

Advanced Usage

curl -X POST http://localhost:5000/consult \
-H "Content-Type: application/json" \
-d '{"agent":"Sergey","consultation_context":"I need to find information about a specific topic.","search_query":"Specific topic keywords"}'