Two tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (get_weather, get_forecast), which is good. However, descriptions lack actionable context for LLM selection. Input schemas use Zod but lack explicit type annotations in the visible schema structure. Output schemas are mentioned but not formally documented. Error handling returns isError flag but lacks recovery guidance. No parameter constraints (enums, ranges) beyond format enum in get_weather. The server is functional but falls short of production-grade tool definition quality.
获取指定地点的详细天气预报(包括未来几天)
获取指定地点的天气信息。支持城市名、邮编、机场代码等。
Output schemas not formally documented. Both tools return structuredContent with 'weather' and 'forecast' fields, but no explicit output schema definition visible in tool registration. LLMs cannot plan downstream operations without knowing return structure.
Error messages lack recovery guidance. Errors return isError=true with generic messages like '获取天气失败' (get weather failed) but do not suggest next steps (e.g., 'Try a different location format' or 'Check if location name is valid').
get_forecast description is vague (45 chars). '获取指定地点的详细天气预报(包括未来几天)' (Get detailed weather forecast for specified location including next few days) does not explain WHEN to use it vs get_weather or what data structure is returned.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 49 | 2025-06-18+ | v2 |
No parameter constraints beyond format enum. location parameter accepts free-form strings with no validation hints (length, format, examples of valid inputs like 'Beijing' vs 'CN' vs postal codes). LLMs may pass invalid location formats.
No pagination or result limits documented. If wttr.in returns large forecast arrays, no limit is enforced or mentioned. Large responses waste tokens and risk context window exhaustion.