Custom Vision Training Client

MCP Server

This project is an MCP (Multi-Agent Conversation Protocol) Server for the given OpenAPI URL - https://api.apis.guru/v2/specs/microsoft.com/cognitiveservices-Training/3.2/openapi.json, auto-generated using AG2's MCP builder.

Prerequisites

• Python 3.9+
• pip and uv

Installation

1. Clone the repository:
   git clone <repository-url> cd mcp-server
2. Install dependencies: The .devcontainer/setup.sh script handles installing dependencies using pip install -e ".[dev]". If you are not using the dev container, you can run this command manually.
   pip install -e ".[dev]"
   Alternatively, you can use uv:
   uv pip install --editable ".[dev]"

Development

This project uses ruff for linting and formatting, mypy for static type checking, and pytest for testing.

Linting and Formatting

To check for linting issues:

To format the code:

These commands are also available via the scripts/lint.sh script.

Static Analysis

To run static analysis (mypy, bandit, semgrep):

This script is also configured as a pre-commit hook in .pre-commit-config.yaml.

Running Tests

To run tests with coverage:

This will run pytest and generate a coverage report. For a combined report and cleanup, you can use:

Pre-commit Hooks

This project uses pre-commit hooks defined in .pre-commit-config.yaml. To install the hooks:

The hooks will run automatically before each commit.

Running the Server

The MCP server can be started using the mcp_server/main.py script. It supports different transport modes (e.g., stdio, sse).

To start the server (e.g., in stdio mode):

The server can be configured using environment variables:

• CONFIG_PATH: Path to a JSON configuration file (e.g., mcp_server/mcp_config.json).
• CONFIG: A JSON string containing the configuration.
• SECURITY: Environment variables for security parameters (e.g., API keys).

Refer to the if __name__ == "__main__": block in mcp_server/main.py for details on how these are loaded.

The tests/test_mcp_server.py file demonstrates how to start and interact with the server programmatically for testing.

Building and Publishing

This project uses Hatch for building and publishing. To build the project:

To publish the project:

These commands are also available via the scripts/publish.sh script.