Build a Weather MCP Tool

Objective

Build a Python class that represents an MCP weather tool, complete with input validation, error handling, and proper response formatting.

Requirements

Create a class WeatherTool that:

1. Defines the tool schema with a get_tool_definition() method that returns:

   {
       "name": "get_weather",
       "description": "Get current weather for a city",
       "inputSchema": {
           "type": "object",
           "properties": {
               "city": {"type": "string", "description": "City name"},
               "units": {"type": "string", "enum": ["celsius", "fahrenheit"]}
           },
           "required": ["city"]
       }
   }
   

2. Validates input with a validate(arguments: dict) -> bool method:

   • Returns True if valid
   • Raises ValueError with message if invalid
   • Must check: city is non-empty string, units (if provided) is valid

3. Executes the tool with execute(arguments: dict) -> dict method:

   • Simulates weather data (use mock data, no real API needed)
   • Returns: {"city": str, "temperature": int, "units": str, "condition": str}
   • Default units is "celsius"

4. Formats response with format_response(result: dict) -> str method:

   • Returns human-readable string like: "Weather in London: 15°C, Partly cloudy"

Example Usage

tool = WeatherTool()

# Get definition
definition = tool.get_tool_definition()
print(definition["name"])  # "get_weather"

# Validate
tool.validate({"city": "London"})  # True
tool.validate({"city": ""})  # Raises ValueError

# Execute
result = tool.execute({"city": "London", "units": "celsius"})
# {"city": "London", "temperature": 15, "units": "celsius", "condition": "Partly cloudy"}

# Format
text = tool.format_response(result)
# "Weather in London: 15°C, Partly cloudy"

Hints

• For mock weather, you can use a simple dict mapping cities to temperatures
• Use °C for celsius and °F for fahrenheit in formatted output
• Validation should be strict but handle optional fields gracefully