Single tool with minimal documentation. Tool name 'get_forecast' follows verb_noun convention (+), but description is generic (18 chars, below 20-char threshold). Input schema present with proper types (latitude, longitude as float) but parameter descriptions are trivial ('Latitude of the location'). No output schema documented, function returns formatted string, not structured data. No error handling guidance. Tool lacks idempotent/destructive hints. Baseline: 194 chars avg description, 72 chars avg param annotation; this server is 50% below baseline.

• Expand tool description to 100-150 chars: 'Fetch current weather forecast for a location in South Korea using KMA (Korea Meteorological Administration) API. Returns temperature, sky condition, precipitation type, humidity, and wind speed. Requires valid latitude/longitude in decimal degrees within Korea bounds.'
• Document output schema as structured object: {temperature_c: number, sky_condition: string (clear|cloudy|overcast), precipitation_type: string, humidity_percent: number, precipitation_mm: string, wind_speed_ms: number}
• Add parameter constraints: latitude description → 'Latitude in decimal degrees (-90 to 90). South is negative. Must be within Korea (33 - 43°N).' longitude description → 'Longitude in decimal degrees (-180 to 180). West is negative. Must be within Korea (124 - 132°E).'
• Add error handling: 'If coordinates are outside Korea, returns error with nearest valid region. If API key is missing/invalid, returns 'Authentication failed, check PUBLIC_DATA_API_SECRET environment variable.' If API is unavailable, returns 'KMA service temporarily unavailable, try again in 1 minute.'
• Return structured JSON object instead of formatted string: enables downstream tool chaining and reduces LLM parsing errors. E.g. {temperature_c: 15.2, sky_condition: 'cloudy', precipitation_type: 'rain', humidity_percent: 72, precipitation_mm: '2.5', wind_speed_ms: 3.1}