home / mcp / local services mcp server

Local Services MCP Server

MCP Server generated by mcp.ag2.ai

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "ag2-mcp-servers-local-services-api": {
      "command": "python",
      "args": [
        "mcp_server/main.py",
        "stdio"
      ],
      "env": {
        "CONFIG": "{...} as JSON string",
        "SECURITY": "security-related parameters",
        "CONFIG_PATH": "path/to/mcp_config.json"
      }
    }
  }
}

You run an MCP (Model Context Protocol) Server to expose a specific API specification for clients to interact with through a standardized MCP interface. This server is designed to host the Local Services API described by the OpenAPI spec, enabling clients to request data and perform actions in a consistent, protocol-driven way. It handles the MCP communication layer, so you can focus on building clients and workflows that consume the API.

How to use

Start the MCP server in stdio mode and connect your MCP client to it. The server exposes its functionality through a local process, which you can launch from your development or deployment environment. Use the client to send MCP requests and receive responses according to your integration needs, such as querying local services data or invoking specific actions defined by the API.

How to install

Prerequisites: Python 3.9+ and a working Python toolchain (pip and uv). Install or set up the development environment, then install dependencies so you can run and test the MCP server locally.

# Prerequisites
python3.9 --version
pip --version
uv --version

# Step 1: clone the project
git clone <repository-url>
cd local_services_mcp

# Step 2: install dependencies (dev environment)
pip install -e ".[dev]"

# Alternative using uvx (editable)
uv pip install --editable ".[dev]"

Additional setup notes

You can pass configuration to the server via environment variables or a JSON configuration file. The example below describes the variables you can provide to customize security and startup behavior.

Available tools

linting

Ruff is used for linting and formatting to ensure clean, consistent code.

static type checking

Mypy performs static type checking to catch type errors before runtime.

testing

Pytest runs tests to validate functionality and behavior.

pre-commit

Pre-commit hooks enforce code quality before each commit.