# incloud-mcp

**Repository Path**: xihadajiang1/incloud-mcp

## Basic Information

- **Project Name**: incloud-mcp
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-08-14
- **Last Updated**: 2025-08-25

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# InCloud MCP Server

A Model Context Protocol (MCP) server that provides seamless integration with InCloud APIs through automatic Swagger-based tool registration.

## Features

- **Automatic Tool Registration**: Automatically discovers and registers API tools from Swagger documentation
- **Intelligent Decision Flow**: Smart routing between specialized tools and generic API flows
- **Dynamic Service Discovery**: Automatically detects available services from Swagger resources
- **Configurable**: Easy configuration through YAML files
- **Extensible**: Modular architecture for easy extension

## Quick Start

### Installation

```bash
pip install incloud-mcp-server
```

### Configuration

Create a `config.yaml` file:

```yaml
swagger:
  base_url: "http://10.180.210.99:30079"
  
services:
  icompute:
    enabled: true
    apis:
      - method: "GET"
        path: "/v1/vhosts"
        description: "List virtual hosts"
  
  istorage:
    enabled: true
    apis:
      - method: "GET"
        path: "/v1/volumes"
        description: "List storage volumes"

# Authentication (optional)
auth:
  type: "bearer"  # or "basic", "api_key"
  token: "your-token-here"
```

### Usage with MCP Client

Add to your MCP client configuration:

```json
{
  "mcpServers": {
    "incloud-mcp-server": {
      "command": "incloud-mcp-server",
      "args": ["--config", "config.yaml"],
      "env": {}
    }
  }
}
```

## Architecture

### Core Components

1. **Swagger Client**: Discovers APIs from Swagger documentation
2. **Tool Generator**: Dynamically creates MCP tools from API specifications
3. **Prompt Understanding**: Provides intelligent guidance for API usage
4. **Common API Tools**: Generic tools for API discovery and execution

### Workflow

1. **Service Discovery**: Fetch available services from `/swagger-resources`
2. **API Discovery**: Get API documentation from `/v2/api-docs` for each service
3. **Tool Registration**: Automatically register MCP tools for configured APIs
4. **Intelligent Routing**: Route user requests to appropriate tools

## Command Line Options

```bash
incloud-mcp-server --help
```

Options:
- `--config`: Path to configuration file (default: config.yaml)
- `--transport`: Transport type (stdio/sse/streamable-http)
- `--port`: Port number for HTTP transport
- `--host`: Host address for HTTP transport

## Development

### Setup Development Environment

```bash
git clone <repository>
cd incloud-mcp
pip install -e ".[dev]"
```

### Running Tests

```bash
pytest
```

### Project Structure

```
src/incloud_mcp_server/
├── __init__.py              # Main entry point
├── server.py                # MCP server implementation
├── config.py                # Configuration management
├── swagger/                 # Swagger integration
│   ├── client.py           # Swagger API client
│   └── tool_generator.py   # Dynamic tool generation
├── tools/                   # MCP tools
│   ├── common_tools.py     # Common API tools
│   └── prompt_understanding.py  # Intelligent guidance
└── static/                  # Static resources
    └── PROMPT_UNDERSTANDING.md  # AI guidance template
```

## License

MIT License