Two tools with basic schemas and descriptions. Both tools have proper input schemas with type constraints (string length, numeric bounds), which is good. Descriptions are present but minimal (18-27 chars), below the 50-200 char LLM-optimized baseline. No output schemas documented. No error handling guidance. Tool names follow verb_noun pattern correctly. Parameters are typed but lack depth in descriptions.

• Expand tool descriptions to 80-150 chars. Example: 'get-alerts' → 'Retrieve active weather alerts for a US state. Returns event type, severity, area, and headline. Use this to check for warnings before planning outdoor activities.'
• Document output schemas in code comments or JSDoc. Specify that get-alerts returns {features: [{properties: {event, areaDesc, severity, status, headline}}]} and get-forecast returns {properties: {periods: [{name, temperature, windSpeed, shortForecast}]}}.
• Add error handling that returns structured error objects with actionable guidance. Instead of null, return {error: 'Invalid state code. Must be 2-letter US state abbreviation (e.g., CA, NY).'}
• Add parameter descriptions for latitude/longitude explaining the coordinate system and what geographic area the forecast covers (e.g., 'Latitude of the location (-90 to 90). Forecast covers a ~2.5 km grid cell around this point.').
• Document result limits and pagination. If alerts can exceed 50 items, add a limit parameter (default 20) and return a total_count field so agents know if results are truncated.