Server has 4 tools with basic schemas and descriptions, but significant gaps limit production readiness. All tools have descriptions (10-50 chars) and input schemas with type definitions, but descriptions are too brief to guide LLM selection effectively. calculate_sum and weather tools are read-only and well-scoped; edit_file is a write operation but lacks error guidance and confirmation patterns. No tool annotations (readOnlyHint, destructiveHint). Output schemas are undocumented. Parameter descriptions exist but are minimal. No pagination, no batch operations, no recovery guidance for failures.

• Expand tool descriptions to 50-200 characters. Example: 'calculate_sum: Add two integers and return the result. Use this for arithmetic operations. Returns an integer.' Current: 'Add two numbers' (16 chars).
• Add tool annotations via FastMCP decorators: @mcp.tool(readOnlyHint=True) for get_alerts/get_forecast; @mcp.tool(destructiveHint=True) for edit_file.
• Document output schemas explicitly. For weather tools, return structured JSON: {event: str, severity: str, description: str} instead of formatted strings.
• Add error recovery guidance to edit_file: 'If file write fails, check file permissions and disk space. Provide a restore_file tool to revert changes.'
• Add parameter constraints: get_alerts state param should validate against US state codes (enum or regex pattern); get_forecast should validate latitude [-90, 90] and longitude [-180, 180].

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

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