Server has 6 tools with basic descriptions and input schemas present. However, descriptions are verbose (200-400+ chars) and lack LLM-optimized clarity on WHEN to use each tool vs. alternatives. Parameter descriptions exist but are minimal (e.g., 'Date in YYYY-MM-DD format'). Output schemas are not formally documented, responses are free-form formatted strings rather than structured JSON objects. Error handling returns plain text messages without recovery guidance. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite all tools being read-only. Naming is reasonable (verb_noun pattern: bazi_calculate, calendar_convert, huangli_query, fortune_daily, jieqi_query, wuxing_analyze) but lacks disambiguation between similar tools (e.g., bazi_calculate vs wuxing_analyze both take birth_date/birth_time, unclear when to use each).
Calculate BaZi (Eight Characters) for fortune telling / 计算生辰八字用于算命
Convert between solar and lunar calendar / 公历农历互转
Get daily fortune and recommendations / 获取每日运势和建议
Query Chinese almanac (Huangli) for a specific date / 查询指定日期的黄历信息
Query 24 solar terms (Jie Qi) for a year / 查询一年的二十四节气
Analyze Wu Xing (Five Elements) from birth info / 根据出生信息分析五行
Output schemas not documented. All tools return free-form formatted strings (markdown-like) instead of structured JSON. LLMs cannot parse or chain results reliably.
Tool descriptions are verbose (200-400+ chars) and lack clarity on WHEN to use each tool. bazi_calculate and wuxing_analyze both accept birth_date/birth_time but their distinction is unclear. No guidance on tool selection.
No tool annotations despite all tools being read-only. Missing readOnlyHint declarations that would help agents understand safety and retry semantics.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | 2026-07-28+ | v2 |
Error handling returns plain text messages without recovery guidance. E.g., 'Error parsing date/time' does not tell the LLM what to do next or what format is expected.
Parameter descriptions are minimal and lack format/constraint details. E.g., 'Birth time in HH:MM format' does not specify if 24-hour or 12-hour, or whether seconds are accepted.