MCP Weather & News Assistant
An MCP server that provides real-time weather data and latest news headlines via custom tools.
README
MCP Weather & News Assistant
An AI assistant project built with the Model Context Protocol (MCP) that connects an LLM to real-world data sources via custom MCP tools.
It provides real-time weather information and latest news headlines by allowing the model to call external APIs through an MCP server.
Features
- Real-time Weather Data: Get current weather conditions (temperature, humidity, wind speed, description) for any city
- Latest News Articles: Fetch top 5 news articles by topic using NewsAPI
- MCP Protocol: Standards-based integration with any MCP-compatible client using FastMCP
- Two Tools:
weather(city)andnews(topic)tools exposed via MCP - Secure: API keys stored in environment variables using python-dotenv
Prerequisites
Before deploying this MCP server, ensure you have:
- Python 3.8+ installed
- API Keys (free):
- An MCP-compatible client such as:
- Claude Desktop
- Continue (VS Code extension)
- Cline
Installation
1. Clone the Repository
git clone https://github.com/ZiliOussema/mcp_weather_and_news_assistant.git
cd mcp_weather_and_news_assistant
2. Create Virtual Environment
# Windows
python -m venv venv
venv\Scripts\activate
# macOS/Linux
python3 -m venv venv
source venv/bin/activate
3. Install Dependencies
pip install -r requirements.txt
4. Set Up Environment Variables
Create a .env file in the project root (you can copy from .env.sample):
# Copy the sample file
cp .env.sample .env
# Then edit .env and add your API keys
Your .env file should contain:
OPENWEATHER_API_KEY=your_openweather_api_key_here
NEWS_API_KEY=your_newsapi_key_here
IMPORTANT: Never commit your .env file to version control!
Finding Your Project Path
Before configuring any MCP client, you need to know your project's full path.
Windows
Open Command Prompt in your project folder and run:
cd
Or use PowerShell:
pwd
macOS/Linux
Open Terminal in your project folder and run:
pwd
Deployment
1. Claude Desktop
Windows
Step 1: Locate the Claude Desktop configuration file
Path: %APPDATA%\Claude\claude_desktop_config.json
Step 2: Open claude_desktop_config.json and add your MCP server:
{
"mcpServers": {
"weather-news": {
"command": "C:\\path\\to\\your\\project\\venv\\Scripts\\python.exe",
"args": [
"src/server.py"
],
"cwd": "C:\\path\\to\\your\\project"
}
}
}
Important Notes:
- Replace
C:\\path\\to\\your\\project\\with your actual project path - Use double backslashes (
\\) in Windows paths - Paths must be absolute (full path from C:)
Step 3: Verify the paths exist
"C:\path\to\your\project\mcp_weather_and_news_assistant\venv\Scripts\python.exe" --version
Step 4: Restart Claude Desktop
- Completely close Claude Desktop
- Check Task Manager to ensure no Claude processes are running
- Reopen Claude Desktop
Step 5: Verify the server is loaded
- Look for a plug icon or "MCP" section in Claude Desktop
- You should see
weatherandnewstools available - Try asking: "What's the weather in Paris?"
macOS
Step 1: Locate the configuration file
~/Library/Application Support/Claude/claude_desktop_config.json
Step 2: Edit the configuration:
nano ~/Library/Application\ Support/Claude/claude_desktop_config.json
Add:
{
"mcpServers": {
"weather-news": {
"command": "/Users/YOUR_USERNAME/mcp_weather_and_news_assistant/venv/bin/python",
"args": [
"src/server.py"
],
"cwd": "/Users/YOUR_USERNAME/mcp_weather_and_news_assistant"
}
}
}
Step 3: Restart Claude Desktop
killall Claude
Then reopen Claude Desktop from Applications.
Linux
Step 1: Configuration location
~/.config/Claude/claude_desktop_config.json
Step 2: Edit configuration:
nano ~/.config/Claude/claude_desktop_config.json
Add:
{
"mcpServers": {
"weather-news": {
"command": "/home/YOUR_USERNAME/mcp_weather_and_news_assistant/venv/bin/python",
"args": [
"src/server.py"
],
"cwd": "/home/YOUR_USERNAME/mcp_weather_and_news_assistant"
}
}
}
Step 3: Restart Claude Desktop
pkill claude
2. Continue (VS Code)
Step 1: Install Continue extension
- Open VS Code
- Go to Extensions (
Ctrl+Shift+X) - Search for "Continue"
- Click Install
Step 2: Open Continue configuration
- Press
Ctrl+Shift+P(Windows/Linux) orCmd+Shift+P(macOS) - Type:
Continue: Open config.json - Press Enter
Step 3: Add MCP server to Continue config
For Windows:
mcpServers:
weather-news:
command: C:\path\to\your\project\venv\Scripts\python.exe
args:
- src/server.py
cwd: C:\path\to\your\project
For macOS/Linux:
mcpServers:
weather-news:
command: /Users/YOUR_USERNAME/mcp_weather_and_news_assistant/venv/bin/python
args:
- src/server.py
cwd: /Users/YOUR_USERNAME/mcp_weather_and_news_assistant
Step 4: Reload VS Code
- Press
Ctrl+Shift+P/Cmd+Shift+P - Type:
Developer: Reload Window - Press Enter
Step 5: Test in Continue
Open Continue sidebar and ask:
What's the weather in New York?
3. Cline
Step 1: Install Cline
- Open VS Code
- Go to Extensions (
Ctrl+Shift+X) - Search for "Cline"
- Click Install
Step 2: Configure Cline MCP settings
- Click the Cline icon in VS Code sidebar
- Go to Cline Settings (gear icon)
- Find "MCP Servers" section
Step 3: Add configuration
Click "Add MCP Server" and fill in:
- Name:
weather-news - Command: Path to your Python executable
- Args: Path to your server.py
Or edit settings.json:
{
"cline.mcpServers": [
{
"name": "weather-news",
"command": "C:\\path\\to\\your\\project\\venv\\Scripts\\python.exe",
"args": [
"src/server.py"
],
"cwd": "C:\\path\\to\\your\\project"
}
]
}
Step 4: Restart Cline or reload VS Code
Usage Examples
Weather Queries
User: What's the weather like in Tokyo?
Response:
The current weather in Tokyo is 18°C with clear skies.
Humidity is at 65% with a light breeze of 5 m/s.
News Queries
User: Show me the latest technology news
Response:
Here are the top 5 tech headlines:
1. New AI breakthrough announced by researchers
Source: TechCrunch | Published: 2026-02-03
2. Tech company releases innovative product
Source: The Verge | Published: 2026-02-03
...
Combined Queries
User: What's the weather in London and any technology news?
Troubleshooting
Server Not Loading in Claude Desktop
-
Check configuration file location
# Windows echo %APPDATA%\Claude\claude_desktop_config.json # macOS echo ~/Library/Application\ Support/Claude/claude_desktop_config.json # Linux echo ~/.config/Claude/claude_desktop_config.json -
Verify JSON syntax at jsonlint.com
-
Check Python path
# Windows "C:\path\to\your\project\venv\Scripts\python.exe" --version # macOS/Linux /path/to/your/project/venv/bin/python --version -
View Claude logs
- Windows:
%APPDATA%\Claude\logs\ - macOS:
~/Library/Logs/Claude/ - Linux:
~/.local/state/Claude/logs/
- Windows:
"Module not found" Error
-
Activate virtual environment and reinstall
# Windows venv\Scripts\activate pip install -r requirements.txt # macOS/Linux source venv/bin/activate pip install -r requirements.txt -
Verify packages are installed
pip list
API Errors (401, 403, 404)
-
Verify API keys in
.envfilecat .env -
Test API keys manually
# Test OpenWeatherMap curl "https://api.openweathermap.org/data/2.5/weather?q=Paris&appid=YOUR_KEY&units=metric" # Test NewsAPI curl "https://newsapi.org/v2/everything?q=technology&apiKey=YOUR_KEY&pageSize=5" -
Check API key limits
- OpenWeatherMap: 60 calls/minute
- NewsAPI: 100 requests/day (free tier)
Tools Not Appearing in Continue/Cline
-
Reload VS Code:
Ctrl+Shift+P→ "Developer: Reload Window" -
Check logs in VS Code Output panel
-
Test server manually
python src/server.py
"Permission Denied" on macOS/Linux
chmod +x server.py
Project Structure
mcp_weather_and_news_assistant/
├── src/
│ ├── __pycache__/
│ ├── tools/
│ │ ├── __pycache__/
│ │ ├── __init__.py
│ │ ├── news.py # News tool implementation
│ │ └── weather.py # Weather tool implementation
│ └── server.py # Main MCP server entry point
├── test/
│ ├── interactive_test.py # Interactive testing script
│ └── test_tools.py # Unit tests for tools
├── venv/ # Virtual environment
├── .env # Environment variables
├── .env.sample # Sample environment file
├── .gitignore
├── requirements.txt # Python dependencies
└── README.md # Documentation
Testing the Server
Manual Testing
# Activate environment
source venv/bin/activate # or venv\Scripts\activate on Windows
# Run server
python src/server.py
Interactive Testing
The project includes an interactive test script to verify your API keys and tools:
# Activate environment
source venv/bin/activate # or venv\Scripts\activate on Windows
# Run interactive test
python test/interactive_test.py
This will test both the weather and news tools with sample queries.
Unit Tests
Run the unit tests to verify tool functionality:
# Run all tests
python -m pytest test/
# Run specific test file
python test/test_tools.py
Security Best Practices
- Never commit API keys - Always use
.envfile - Add
.envto.gitignore - Rotate keys regularly
- Use read-only keys when possible
MCP Weather & News Assistant - A portfolio project demonstrating Model Context Protocol integration
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。