twexam-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_questionsB | 全文搜尋歷屆考題(匹配題幹/選項/擬答)。 |
| get_questionA | 以 qid(年-考試-科目-題號)取得單題結構化內容。 |
| list_examsA | 列出可查的考試別與年度範圍。 |
| list_subjectsB | 列出科目(可選 exam_code 篩選)。 |
| get_exam_paperB | 取整份試卷(某年·某考試·某科目全部題目)。 |
| get_answer_keyB | 取某份試卷的測驗題標準答案(題號→答案)。 |
| get_model_answerB | 取申論題 AI 擬答(含免責聲明)。 |
| search_by_statuteB | 按法條反查考過哪些題。 |
| get_statute_frequencyC | 法條考頻統計(可選 exam_code)。 |
| random_practiceC | 依條件抽題練習(hide_answer 可隱藏答案/擬答)。 |
| get_exam_mapA | 考點地圖:各科目層級與核心考點(trial='sl1'/'sl2'/None=全部)。 |
| get_topic_distributionA | 考點熱度排行:各 (子科目, 考點) 考過幾題(q_type='mcq'/'essay',exam_code='sl1'/'sl2')。 |
| practice_by_topicB | 依考點抽題練習(topic_point 為必填,如「抵押權(普通/最高限額)」)。 |
| essay_exam_by_topicA | 考點申論題卷(模擬考):把某考點或某子科目的所有申論題一次考出來,預設直接附 AI 擬答。 topic_point 或 topic_subject 至少給一個;show_answer=False 可先自己作答再看擬答。 |
| record_answerB | 記錄一次作答並自動批改+更新間隔重複排程。MCQ 自動對答案;申論可傳 self_correct 自評。 |
| get_grading_rubricA | 申論批改評分表(學生寫完申論作答後呼叫):一次備齊批改該題所需素材——爭點 checklist +學說對立+實務字號+🔍辨識訊號/📐前置觀念+滿分擬答+五維評分準則。供你逐爭點對照 學生作答給形成性回饋(爭點漏抓/要件未涵攝/學說未選邊/字號漏引/架構),不打硬分數。 |
| get_error_diagnosisA | 錯誤類型診斷(不只正確率,要『為什麼錯』):回傳每筆答錯題帶——你反覆選的錯選項+正解 +逐項詳解+該題錯幾次+考點,並標『同考點群聚』。據此歸納誤解類型:①觀念混淆(把A當B) ②掉陷阱(選最誘人的錯選項)③粗心(偶發)。wrong_times≥2 或群聚=系統性,優先補。 |
| get_weak_topicsC | 弱點地圖:依個人作答正確率,由弱到強列出各考點(含答對率)。 |
| practice_weakA | 弱點練習:優先出「今天到期複習」與「最弱考點」的題目(hide_answer 預設隱藏答案)。 |
| get_progressC | 學習總覽:總作答數、答對率、已練題數、今天到期複習數。 |
| reset_progressA | 清空所有作答記錄與複習排程(重新開始)。 |
| get_readinessC | 考試就緒度:依考點頻率加權推估分數、覆蓋率、最拖分考點、每日覆蓋進度(target=及格參考線)。 |
| get_study_planA | 讀書計畫引擎(綁考試日的處方):把「弱點×考點頻率」排成攻擊順序,依剩餘天數排出分相日程 (掃弱點→申論模擬→衝刺複習)+今日該做什麼+進度夠不夠的誠實判斷。days_remaining 請用 使用者的考試日期減今天算出。非及格保證,作答數據越多越準。 |
| get_topic_primerC | 考點重點提示:做題前必讀的核心法條/常考判決釋字/學說對立/易錯陷阱。 |
| get_issue_distributionB | 爭點熱度排行:申論真實考點(學說/實務交鋒點)考過幾題,由多到少。可選子科目篩選。 |
| search_by_issueA | 依爭點找題:某爭點(如「不能未遂之判斷標準」)考過哪幾題,附各題的學說對立與實務見解。 |
| get_issuesA | 取某申論題拆出的所有爭點:爭點名+學說對立+實務見解(判例/決議/釋字/憲判字號)。 |
| get_issue_primerA | 申論爭點重點包(做題前必讀):🔍辨識訊號(怎麼從事實認出此爭點)+📐前置觀念 (要先懂的定義/法理)+考點重點。issue 可給標準爭點名或原始爭點字串(自動正規化)。 |
| get_issue_chainB | 爭點脈絡圖:一題的多個爭點不是孤立的——回傳它們的『先決問題』鏈(誰是誰的先決問題、 事實哪個轉折引爆下一個爭點)。破解「孤立考點背熟、遇綜合題卻看不出脈絡」。帶使用者解綜合 申論前先看脈絡;單爭點題或尚未編脈絡者 chain 為空。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 29 tools
Most tools have clearly distinct purposes, but a few pairs could be confused, such as get_issue_primer vs get_topic_primer and get_issue_distribution vs get_topic_distribution, which both offer preparatory/statistical content centered on either issues or topics. Descriptions are detailed enough to resolve most ambiguities, but the overlap is notable.
The vast majority follow snake_case verb_noun conventions (e.g., get_readiness, list_subjects, search_questions, practice_by_topic). Minor deviations like random_practice and essay_exam_by_topic break the pattern slightly, but overall the naming is predictable and readable.
At 29 tools, the surface is on the heavy side, but each tool serves a distinct function within a comprehensive exam preparation workflow—from question retrieval and practice to analytics, grading, and study planning. The count is justified by the breadth of the domain, though consolidation of a few overlapping utilities could make it tighter.
The tool set covers the full exam prep lifecycle: exam and question discovery, practice in multiple modes, answer recording and grading, progress tracking, weakness analysis, study planning, and detailed issue breakdowns. There are no obvious dead ends or missing core operations for the stated purpose.