MCP Server - Z.ai Integration

A Model Context Protocol (MCP) server that provides filesystem access and Z.ai GLM-4 model integration for MCP clients.

Features

Filesystem Access: Read/write files in the parent repository
Z.ai Integration: Use local Z.ai API key with GLM-4 models for code generation
Submodule Detection: Automatically detects submodule environment and sets working root to parent directory

Installation as Git Submodule

cd your-parent-project
git submodule add https://github.com/your-username/mcp-server.git mcp-server
cd mcp-server
pip install -r requirements.txt
cp .env.example .env

Edit .env and add your Z.ai API key:

ZAI_API_KEY=your_actual_api_key

MCP Client Configuration

Claude Desktop

Add to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "zai-server": {
      "command": "python",
      "args": ["mcp-server/src/main.py"],
      "env": {
        "ZAI_API_KEY": "your_api_key_here"
      }
    }
  }
}

Tools

`read_file`

Read the contents of a file relative to the parent repository root.

`write_file`

Write content to a file relative to the parent repository root.

`zai_generate`

Generate code or perform reasoning tasks using Z.ai GLM-4 models.

Parameters:

prompt (string): The prompt to send to the model
model (string, optional): Model identifier (default: "glm-4")
temperature (number, optional): Sampling temperature (default: 0.7)
max_tokens (number, optional): Maximum tokens to generate (default: 2000)

Usage Example

Once configured, you can use the tools from your MCP client:

User: Read the main.py file
Assistant: [Uses read_file tool]

User: Generate a sorting algorithm using Z.ai
Assistant: [Uses zai_generate tool with prompt about sorting algorithms]

License

MIT

mcp