Two tools with basic verb-noun naming (get_alerts, get_forecast). Both have short descriptions (34-40 chars, below the 50-200 char baseline for LLM optimization). Input schemas are present with typed parameters and descriptions, but output is unstructured plain text. No error handling guidance, no pagination, no structured response schema. Tools are read-only but lack explicit risk annotations. Naming is clear but descriptions are minimal.

• Expand tool descriptions to 80-150 chars. Example for get_alerts: 'Fetch active weather alerts for a US state. Returns event type, severity, affected area, and instructions. Use this to check for warnings before planning outdoor activities.'
• Return structured JSON objects instead of formatted strings. get_alerts should return {alerts: [{event, area, severity, description, instructions}], count, state}. get_forecast should return {periods: [{name, temperature, windSpeed, windDirection, forecast}], location, generatedTime}.
• Add error classification: distinguish between 'invalid_state_code' (user-fixable: 'State code must be 2 uppercase letters'), 'api_unavailable' (retryable), and 'no_data' (informational). Include recovery hints.
• Add optional limit and offset parameters to get_alerts. Document: 'Returns up to 50 alerts per request. Use offset for pagination. Default limit=20.'
• Document the 5-period limit in get_forecast description and expose it as an optional parameter (1-10, default=5) so users can request extended forecasts.