SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsmaster-thesis-review

master-thesis-review

Writing & research

View repository

Updated 10-15 18:46

About this skill

Master Thesis Review (硕士论文评阅意见)

Overview

This skill generates a master's thesis review report (硕士学位论文评阅意见) for a given thesis. The user provides the thesis file and a score (out of 100). The skill reads the thesis, then generates two sections calibrated to the score level.

Input:

  • A project folder containing the thesis (*.pdf, *.tex, or *.docx)
  • A score from the user (integer, 0–100)

Output: {original-filename}-评阅意见.doc saved in the project folder

Language Rule

  • Communicate with the user in Chinese.
  • Write the review report entirely in Chinese.

Workflow

Phase 1: Initialization

  1. Receive folder path from user (via $ARGUMENTS or ask: "请提供论文文件夹路径").
  2. Scan the folder using Glob to locate the thesis file (*.pdf, *.tex, *.docx).
  3. Read the thesis in full. For large PDFs, use page ranges. Understand:
    • Thesis title (title)
    • Research question and topic background
    • Scope of literature review
    • Research methods and data
    • Main conclusions and innovations
    • Writing structure and standardization

Phase 2: Configuration

Use one AskUserQuestion call with one question:

Question — Score (Score):

  • Ask: "请输入您给该论文的分数（满分100分）"
  • Options: 90–100 分 / 80–89 分 / 70–79 分 / 60–69 分
  • User can type exact score via "Other" (e.g., 85)

Record the exact score. If the user selects a range, use the midpoint (e.g., 90–100 → 95).

Phase 3: Generate Review

Generate two sections based on the thesis content and calibrated to the score.

Score Calibration Guide

The score determines the tone, praise-to-criticism ratio, and severity of issues raised:

90–100 分 (优秀)

  • Academic evaluation: Highly affirm the significance and innovativeness of the topic. Fully acknowledge the breadth and depth of the literature mastered. Affirm the reliability and standardization of the data/methods. Tone is enthusiastic and positive.
  • Shortcomings: Only raise minor, non-substantive improvement suggestions (e.g., individual expressions could be more concise, a certain point could be further discussed). Tone is gentle, mainly "suggestions" rather than "problems."

80–89 分 (良好)

  • Academic evaluation: Affirm the practical significance and academic value of the topic, and acknowledge the reasonableness of the research methods. While affirming, may imply that certain aspects still have room for improvement. Tone is positive but not overly exaggerated.
  • Shortcomings: Point out 2–3 substantive but non-fatal issues (e.g., literature coverage could be broadened, certain arguments could be strengthened, a certain step in data processing could be improved). Tone is objective and constructive.

70–79 分 (中等)

  • Academic evaluation: Affirm that the topic has some significance, but the wording is relatively plain. Point out that literature, methods, data, etc. are basically acceptable but have obvious deficiencies. Tone is neutral and somewhat conservative.
  • Shortcomings: Point out 3–4 relatively obvious issues (e.g., research methods are not rigorous enough, depth of literature review is insufficient, data sources or processing are questionable, logical chain is not complete enough). Tone is direct but still constructive.

60–69 分 (及格)

  • Academic evaluation: Only acknowledge that the topic has some reference value. Give relatively low evaluations of literature mastery, method application, writing standardization, etc. Tone is restrained, with few positive statements.
  • Shortcomings: Point out multiple relatively serious issues (e.g., logical loopholes in the core argument, doubtful data reliability, weak literature review, insufficient innovativeness, non-standard writing). Tone is serious, emphasizing the need for substantial revision.

Report Structure

Section 1: 对学位论文的学术评语

Requirements:

  • 240 字（中文字符，±10%）
  • Do not use bullet points, write as one complete paragraph
  • Must cover the following five aspects (naturally integrated, do not list them one by one):
    1. Significance of the topic
    2. Mastery of literature and materials
    3. Reliability of the materials and data used
    4. Writing standardization and logicality
    5. Innovations of the thesis
  • Adjust the ratio of praise to criticism and the strength of wording according to the score band
  • Must be based on the paper'sactual content, citing specific research topics, methods, variables, findings, etc.

Section 2: Shortcomings and suggestions of the paper

Requirements:

  • 300 characters(Chinese characters, ±10%)
  • Do not use bullet points, write as one complete paragraph
  • Clearly point out the specific problems and shortcomings in the paper
  • Propose actionable revision suggestions
  • The number and severity of the problems should match the score band (high score → few and minor; low score → many and severe)
  • Must be based on the paper'sactual content, point out problems in specific chapters, tables, and arguments

Phase 4: Output

  1. Determine output filename: Strip extension from thesis filename, append -评阅意见.doc. Example: 张三_硕士论文.pdf → 张三_硕士论文-评阅意见.doc.

  2. Assemble the report with this structure:

对学位论文的学术评语

[240字学术评语段落]

论文的不足之处和建议

[300字不足与建议段落]

建议成绩：XX 分

  1. Generate .doc using python-docx (preferred) or pandoc. Font: 宋体 (SimSun), size 12pt, 1.5x line spacing.

  2. Display the full report content to the user.

  3. Tell the user: "The review comments have been generated and saved to {filename}-评阅意见.doc。"

Critical Rules

  1. Strictly adhere to the character count. Academic comments 240 characters (±10%), shortcomings and suggestions 300 characters (±10%). Here "characters" refers to the number of Chinese characters.
  2. Do not use bullet points. Both parts must be coherent paragraphs; numbering, sequence numbers, and bullet points are prohibited.
  3. Base on actual content. All evaluations must target the paper's real content (title, methods, data, conclusions), and must not be vague and general.
  4. Score calibration. The ratio of praise to criticism, the strength of wording, and the severity of the problems must be consistent with the given score. Comments for a 90 score must not be harsher than those for a 70 score.
  5. Academic style. Use formal academic review language, consistent with the conventions and format of Chinese master's thesis review.
  6. Read before writing. Must read the paper completely before writing the comments; do not generate based on partial content.
  7. Do not use an AI tone. Avoid clichés such as "in short", "in summary", and "it is worth affirming that". The language must be natural and professional, like that written by a real review expert.