Server defines 5 weather tools with complete JSON schemas and basic descriptions. All tools follow verb_noun naming (get_*). Schemas include proper type constraints (latitude/longitude bounds, altitude minimum). However, descriptions are generic and lack context for LLM selection, they do not explain WHEN to use each tool or how they differ. Parameter descriptions are minimal (e.g., 'Latitude coordinate (max 4 decimals)' lacks actionable guidance). No output schemas documented. Error handling is present but not visible in tool definitions. Missing tool annotations (readOnlyHint). Composition is reasonable, tools are single-purpose and chainable via lat/lon parameters.

• Expand tool descriptions to 100-150 chars each, explaining WHEN to use each forecast variant. Example: 'get_hourly_forecast: Returns weather predictions for the next 1-48 hours in 1-hour intervals. Use for short-term planning (today/tomorrow). Includes temperature, wind, precipitation per hour.'
• Document output schemas for each tool. Specify returned fields (e.g., temperature, wind_speed, precipitation, timestamp) with types and units. Include example response structure.
• Add readOnlyHint: true to all tool definitions to signal safety and enable agent optimizations.
• Enhance parameter descriptions with units and precision guidance. Example: 'latitude: Decimal degrees (-90 to 90). Use 2-4 decimals for city-level precision (~1-10 km accuracy).'
• Add a discovery tool or clarify in descriptions: 'Call get_current_weather for now, get_hourly_forecast for next 48 hours, get_daily_forecast for 7-day outlook, get_detailed_forecast for all available meteorological data.'