Two tools with basic schemas and minimal descriptions. get_alerts and get_forecast both have valid JSON Schema with required parameters and type definitions, but descriptions are extremely terse (under 50 chars each). No parameter descriptions provided, LLMs cannot infer what 'state' means or why latitude/longitude are needed. No output schemas documented. No error handling guidance. No tool annotations. Naming is acceptable (verb_noun pattern) but lacks context about prerequisites, return structure, or when to use each tool.

• Add parameter descriptions: 'state: Two-digit US state code (e.g., CA, NY, TX). Used to filter weather alerts for that state.' and 'latitude: Latitude in decimal degrees (-90 to 90). Precision to 2 decimal places recommended.' and 'longitude: Longitude in decimal degrees (-180 to 180).'
• Expand tool descriptions to 100-200 chars. Example: 'get_alerts: Retrieve active weather alerts (severe thunderstorms, tornado warnings, etc.) for a US state. Returns a list of alerts with severity, type, and affected areas. Call this to check for dangerous conditions before planning outdoor activities.'
• Document output schemas. Example: 'Returns: {alerts: [{id, type, severity, description, affected_area, issued_at, expires_at}], total_count}' for get_alerts.
• Add error handling guidance: 'If state code is invalid, returns error with list of valid codes. If coordinates are outside US, returns empty result with message.'
• Add tool annotations to schema: include 'readOnlyHint: true' and 'idempotentHint: true' for both tools to signal they are safe to retry.

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry