360 Secure SkillHub
360 安全技能中心

Social Science Journal Abstract Polisher

根据输入的社会科学摘要,按背景、方法、发现、启示四段检查逻辑衔接,执行 APA 第7版规范、学术语体与简洁性润色,输出完整摘要、3至5条改动说明和最终字数;不新增数据或改变研究设计。

terrycarter1985
terrycarter1985
热度 24.91
最近更新 2026-05-12 19:11:47
分类:效率依赖:纯提示词应用场景:校对润色 / 研究

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add social-science-journal-abstract-polisher
Metadata
name
social-science-journal-abstract-polisher
description
Reusable prompt for refining social science academic abstracts to align with peer-reviewed journal requirements including APA 7th edition compliance. Use when polishing, editing, or improving an abstract for a social science journal submission, or when asked to make an abstract conform to APA style and peer-review standards.

Social Science Journal Abstract Polisher

Task

Refine the provided social science abstract for peer-reviewed journal submission without altering the core research content.

Procedure

  1. Read the input abstract in full. Identify the four functional segments (Background, Method, Findings, Implications). If any segment is missing or conflated, flag it and reconstruct from context.

  2. APA 7th Edition Compliance

    • Abstract word count: 150–250 words (unless a specific journal requires otherwise).
    • Use past tense or present perfect for methods and results; present tense for implications and general conclusions.
    • No citations, footnotes, or abbreviations without prior definition within the abstract.
    • Numerals for numbers 10 and above; words for zero through nine (except measurements, ages, percentages, and sample sizes).
    • Active voice preferred; passive only where the action, not the actor, is the focus.
    • No first-person pronouns; rephrase to maintain an objective register.
  3. Logical Cohesion Across Segments

    • Background → Method: The method must clearly follow from the research gap or question stated in the background. Add a transitional phrase if the jump feels abrupt.
    • Method → Findings: Findings should directly answer the method's inquiry. Ensure the dependent variable(s) in the method are the same ones reported in the findings.
    • Findings → Implications: Implications must stem from reported findings, not from aspirational claims. If an implication overreaches, scale it back; if a finding lacks an implication, add one grounded in the data.
    • Use lexical cohesion (repetition of key terms, synonym chains) across segments rather than introducing new terminology mid-abstract.
  4. Academic Register

    • Replace informal or conversational phrasing with formal academic equivalents.
    • Eliminate hedging language that undermines clarity ("somewhat," "arguably," "it could be said that") unless the evidence genuinely warrants uncertainty.
    • Remove filler phrases ("It is important to note that," "In this study, we aimed to," "The purpose of this paper is to").
  5. Conciseness

    • Delete redundant modifiers, tautological constructions, and restatements.
    • Merge sentences that convey overlapping information.
    • Preserve all substantive content; cut only rhetorical padding.
  6. Output Format

    • Return the polished abstract as a single continuous paragraph (or structured format if the target journal specifies labeled sections).
    • Append a brief change log listing the three to five most significant revisions made (e.g., "Removed first-person pronoun," "Added transitional phrase between method and findings").
    • Confirm final word count.

Constraints

  • Do not introduce new empirical claims, data points, or theoretical frameworks not present in the input.
  • Do not alter the study's design, sample, or findings.
  • If the input abstract is fundamentally flawed (e.g., missing findings entirely), note the deficiency and revise what is available rather than fabricating content.