Public leaderboard

Public assessment

cantian-ai/bazi-mcp (bazi-mcp)

bazi-mcp · v0.0.1 · scanned

What changed in the harness

Selection accuracy 98→100, token cost up 14%, unconfirmed writes 100%→100%.

Category breakdown

Where the score comes from.

Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.

01Safety

30.0 / 30

30.0 out of 30
02Legibility

28.4 / 30

28.4 out of 30
03Economics

20.0 / 20

20.0 out of 20
04Discoverability

19.2 / 20

19.2 out of 20

Highest-impact fix

Estimated gain +1 point

Rewrite 3 underspecified tool descriptions

State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks.

Description evidence

Defects and rewrites.

3 defects found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.

Tool Defect types Suggested rewrite
getBaziDetail
no_return_description
根据公历(solarDatetime)或农历(lunarDatetime)时间(二者必须传入且只能二选一)与性别(0为女、1为男)计算八字,并按 eightCharProviderSect 指定的早晚子时规则确定23:00-23:59的日柱归属当天还是次日。返回该时间与性别对应的八字详情,即年柱、月柱、日柱、时柱四柱信息。
getSolarTimes
no_return_description
根据给定的八字(按年柱、月柱、日柱、时柱顺序以空格分隔,例如 戊寅 己未 己卯 辛未)反查其对应的公历时间。返回一个公历时间字符串列表,每个时间格式为 YYYY-MM-DD hh:mm:ss,例如 1998-07-31 14:00:00。
getChineseCalendar
no_return_description
查询指定公历时间(不传则默认今天)对应的黄历信息。传入ISO格式的公历时间(如2008-03-01T13:00:00+08:00)以指定查询日期,并返回该日期对应的中国农历黄历内容。

Selection evidence

Confusable tool pairs.

2 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.

Tool A Tool B Confidence Why they collide
getBaziDetail getSolarTimes medium The two are functional inverses: getBaziDetail converts time→bazi, getSolarTimes converts bazi→time. A natural task phrased as '根据八字算时间' or '把这个时间换成八字' could make an agent pick the wrong direction, since both descriptions are mirror images of each other.
getBaziDetail getChineseCalendar low Both tools accept a solarDatetime in identical ISO format and belong to the same Chinese calendar/fortune-telling domain. A vague task like '查一下这天适不适合结婚/看这个日子的信息' could make an agent pick the almanac when bazi was intended or vice versa, though the outputs (八字 vs 黄历) are described distinctly.

Compare the field

One score is useful.
The evidence makes it actionable.

Back to the leaderboard