AI Service Platform

A cloud-ready service platform for AI-powered tool execution with Model Context Protocol (MCP) integration.

Key Features

• Cloud Native Architecture

  • REST API endpoints for all operations
  • Stateless design with persistent tool configuration
  • Horizontal scaling support

• Unified Tool Gateway

  • Automatic discovery of MCP tools in servers/ directory

Cloud Deployment

Prerequisites

• Python 3.10+

API Usage

Endpoints

List Available Tools

GET /tools

Execute Natural Language Query

POST /query
{
  "query": "5+5",
}

Development Setup

1. Clone the repository:

git clone https://github.com/Daniel1989/mcp-server-cloud.git
cd mcp-server-cloud

2. Set up virtual environment:

python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

3. Start development server:

FLASK_DEBUG=1 python flask.py

Resources

1. python-sdk. https://github.com/modelcontextprotocol/python-sdk
2. cline's prompt -- how to ask ai to select mcp server. https://github.com/cline/cline/blob/main/src/core/prompts/system.ts