first install dependencies using uv package
uv add requirements.txt
you can use
python -m pip install requirements.txt
Basic MCP Project Setup
This repository contains the basic setup steps for a project utilizing uv for dependency management and environment setup, and configuring tools to be used within an MCP (Multi-tool Control Panel) environment, potentially like Claude Desktop.
Prerequisites
- Python 3.7+ (recommended)
- Network access to download
uvand project dependencies.
Installation
This section guides you through setting up the uv tool and initializing your project environment.
1. Install uv
uv is used for managing Python environments and dependencies efficiently. Choose the command corresponding to your operating system.
Windows (PowerShell):
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
macOS / Linux (Bash):
curl -LsSf https://astral.sh/uv/install.sh | sh
After running the installation command, it's crucial to restart your terminal or command prompt to ensure the uv command is recognized in your PATH.
2. Project Setup
Now, let's create the project directory, set up the virtual environment, install necessary dependencies, and create the main server file.
macOS / Linux:
# Create a new directory for our project (e.g., 'weather')
uv init weather
# Navigate into the project directory
cd weather
# Create a virtual environment inside the project folder
uv venv
# Activate the virtual environment
source .venv/bin/activate
# Install project dependencies.
# mcp[cli] requires extra dependencies for the command-line interface.
uv add "mcp[cli]" httpx
# Create our server file (where your application code will reside)
touch weather.py
Windows (Command Prompt or PowerShell):
# Create a new directory for our project (e.g., 'weather')
uv init weather
# Navigate into the project directory
cd weather
# Create a virtual environment inside the project folder
uv venv
# Activate the virtual environment
.venv\Scripts\activate
# Install project dependencies.
# mcp[cli] requires extra dependencies for the command-line interface.
uv add mcp[cli] httpx
# Create our server file (where your application code will reside)
new-item weather.py
At this point, you have a project directory (j), a dedicated virtual environment (.venv), installed libraries (mcp[cli], httpx), and an empty file (``) where you can add your application logic.
Configuration (for Claude Desktop MCP)
If you are using this project with Claude Desktop or a similar MCP tool, you will need to configure it to recognize and run your tools. Below is an example of a configuration structure that defines how different tools are launched.
Note: The exact location and format of the MCP configuration file depend on the specific MCP software you are using (e.g., Claude Desktop's settings). The following is the content of a potential configuration section:
{
"mcpServers": {
"product_data_tools": {
"command": "uv",
"args": [
"--directory",
"C:\\Users\\asus4\\OneDrive\\Bureau\\mcp-Project\\project\\mcp-project",
"run",
"server.py"
]
},
"Bright Data": {
"command": "npx",
"args": ["@brightdata/mcp"],
"env": {
"API_TOKEN": "api_token",
"WEB_UNLOCKER_ZONE": "unlocker_zone",
"BROWSER_ZONE":"your BROWSER_ZONE"
}
},
"actors-mcp-server": {
"command": "npx",
"args": [
"-y",
"@apify/actors-mcp-server",
"--actors",
"autofacts/shopify"
],
"env": {
"APIFY_TOKEN": "api_token"
}
}
}
}
product_data_tools: This entry defines a tool that runs your Python server file (server.py, though in our setup we createdweather.py- you'll need to adjust theargsif you stick toweather.py). It usesuv runto execute the file within the virtual environment managed byuv. Remember to update the--directorypath to the actual location of your project directory (weather) on your system.Bright Data: Configures a tool that runs a Bright Data MCP server usingnpx. Requires specific environment variables for authentication and zone selection.actors-mcp-server: Configures a tool that runs an Apify Actors MCP server usingnpx. Requires an API token.
After adding or updating the configuration file within your MCP software, save the file, and restart Claude for Desktop (or your specific MCP tool) for the changes to take effect.
Usage
Once the environment is set up and the MCP configuration is applied and the MCP tool is restarted, you should be able to interact with the configured tools (product_data_tools, Bright Data, actors-mcp-server) directly through the interface of your MCP software (e.g., Claude Desktop).
The specific way you "use" each tool depends on its implementation. For product_data_tools, you would typically interact with the API or functionality provided by the server.py file you created.
Visuals / Examples (Placeholder)
An image or screenshot demonstrating how to trigger or interact with the configured tools within the Claude Desktop (or your chosen MCP software) interface would be helpful here.
[Insert URL of relevant image here, or delete this section if no image is provided]
run servers
uv run server.py
uv run mcp_data_server.py
This README provides a clear, step-by-step guide for setting up the project and configuring it for use with an MCP tool based on your input. Remember to fill in any placeholders like specific paths or API tokens.