# MCP SuperiorAPIs Local
This project is a Python-based **MCP Server** that dynamically retrieves plugin definitions from **SuperiorAPIs** and auto-generates MCP tool functions based on their OpenAPI schemas.
It operates in `stdio` mode, making it ideal for local development and testing with AI clients.
If you need to integrate using HTTP or SSE protocols, please refer to: [CTeaminfo/mcp_superiorapis_remote](https://github.com/CTeaminfo/mcp_superiorapis_remote)
## 📂 Project Structure
```
mcp_superiorapis_local/
├── src/mcp_superiorapis_local/ # Main program
│ ├── __init__.py # Package initialization
│ └── server.py # MCP server implementation
├── tests/ # Test files
├── pyproject.toml # Project config & dependencies
├── uv.lock # Locked dependencies
└── README.md # Project documentation (this file)
```
## 🚀 Quick Start
### 1. Environment Preparation
**Prerequisites:**
- Python 3.13+
- Superior APIs Token ([How to get](https://superiorapis-creator.cteam.com.tw))
### 2. Clone the Project
```bash
# Using HTTPS
git clone https://github.com/CTeaminfo/mcp_superiorapis_local.git
# Using SSH
git clone git@github.com:CTeaminfo/mcp_superiorapis_local.git
cd mcp_superiorapis_local
```
### 3. Install uv (if not installed)
```bash
# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
# Windows
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
# Or use pip
pip install uv
```
### 4. Install Dependencies
```bash
# Create virtual environment
uv venv --python 3.13
# Install dependencies
uv sync
# Or use pip
pip install -e .
```
### 5. Configure Environment Variables
```bash
# Set your Superior APIs token
export TOKEN=your_superior_apis_token_here
# Windows CMD
set TOKEN=your_superior_apis_token_here
```
**Token Authentication Instructions:**
- Get your token from [Superior APIs](https://superiorapis-creator.cteam.com.tw)
- Set the TOKEN environment variable before running the server
### 6. Start the Server
```bash
python -m mcp_superiorapis_local
or
python src/mcp_superiorapis_local/server.py
```
### 7. Verify Deployment
The server will:
1. Fetch plugin data from SuperiorAPIs
2. Dynamically generate MCP tool functions
3. Register the tools
4. Start the MCP server in stdio mode
## 🔌 MCP Client Integration
### With uvx on Pip
Configure MCP server with uvx on pip(No need to download source code):
```json
{
"mcpServers": {
"mcp_superiorapis_local": {
"command": "uvx",
"args": [
"mcp-superiorapis" // https://pypi.org/project/mcp-superiorapis/
],
"env": {
"TOKEN": "your_superior_apis_token_here"
}
}
}
}
```
### Local Mode
```json
{
"mcp_superiorapis_local": {
"command": "uv",
"args": [
"run",
"--directory",
"/path/to/mcp_superiorapis_local",
"python",
"-m",
"mcp_superiorapis_local"
],
"env": {
"TOKEN": "your_superior_apis_token_here"
}
}
}
```
## 🔧 Startup Steps
```cmd
# 1. Navigate to the project directory
cd mcp_superiorapis_local
# 2. Activate the virtual environment
.venv\Scripts\activate
# 3. Set environment variable
set TOKEN=your_superior_apis_token_here
# 4. Run the project
python -m mcp_superiorapis_local
or
python src/mcp_superiorapis_local/server.py
```
**Note：**
- Dependencies only need to be installed once (using pip install -e . or uv sync)
- After a reboot, you only need to activate the virtual environment and set the environment variable
- Once the virtual environment is active, the command prompt will show a (venv) prefix
## 🔗 Related Links
- [Superior APIs](https://superiorapis-creator.cteam.com.tw) - Obtain your API Token
- [MCP SuperiorAPIs Remote](https://github.com/CTeaminfo/mcp_superiorapis_remote) - HTTP/SSE version
- [MCP Protocol](https://modelcontextprotocol.io/) - Official Model Context Protocol documentatio
## MCPHub Certification
This project is officially certified by [MCPHub](https://mcphub.com).
View this project on MCPHub:
🔗 https://mcphub.com/mcp-servers/CTeaminfo/mcp-superiorapis