home / mcp / resource settings mcp server
MCP Server generated by mcp.ag2.ai
Configuration
View docs{
"mcpServers": {
"ag2-mcp-servers-resource-settings-api": {
"command": "python",
"args": [
"mcp_server/main.py",
"stdio"
],
"env": {
"CONFIG": "{... JSON configuration ...}",
"SECURITY": "YOUR_API_KEYS_OR_SECURITY_VARS",
"CONFIG_PATH": "mcp_server/mcp_config.json"
}
}
}
}You run an MCP Server to expose a specific OpenAPI-based service so clients can interact with it using the MCP protocol. This server lets you start a local process (stdio) that speaks MCP transport, enabling fast development, testing, and integration with MCP clients.
Start the MCP server in stdio mode and connect your MCP client to the local process. The server accepts its configuration from environment variables or a JSON config file. Use the client to request data, run actions, or query the OpenAPI-described resources through the MCP interface. You can switch transport modes later if needed, but stdio is the simplest for local development.
Prerequisites include Python 3.9 or newer, and the tools pip and uv for installation and running the development environment.
Install dependencies and prepare the development environment by installing the package in editable/development mode. If you are using a development container, the same steps apply inside that environment.
>> Install dependencies locally
pip install -e ".[dev]"
>> Alternative: use uv to run the installation command interactively
uv pip install --editable ".[dev]"You can start the MCP server in stdio mode using Python and the main script path. The environment can supply configuration details, including the path to a JSON config file, a JSON string with the configuration, and security-related environment variables.
>>> Start the server in stdio mode using the script path
python mcp_server/main.py stdio