Claude Skill
labarba/sciwrite
SciWrite 是一个基于 Kristin Sainani 博士《科学写作》方法论的 Claude Skill,用于 AI 辅助手稿写作与审阅,帮助提升科学写作的清晰度和结构。
概览
仓库信息
安装这个 Skill
git clone https://github.com/labarba/sciwrite.gitRegistry 信息
项目简介
SciWrite 是一个基于 Kristin Sainani 博士《科学写作》方法论的 Claude Skill,用于辅助手稿写作与审阅,帮助研究人员提升科学写作的清晰度、简洁性和结构。
Agent Skill for AI-assisted manuscript writing review, based on Dr. Kristin Sainani's "Writing in the Sciences" methodology.
要点
- 基于 Kristin Sainani 博士《科学写作》方法论
- AI 辅助手稿写作与审阅
- 注重清晰度、简洁性和结构
- 专为研究人员和学者设计
使用场景
- 提升科学手稿的清晰度
- 审阅和编辑学术论文
- 学习有效的科学写作技巧
README 摘要
# SciWrite — Manuscript Writing Review An [Agent Skill](https://agentskills.io) that reviews scientific and engineering manuscripts for writing quality, applying the methodology from Dr. Kristin Sainani's [*Writing in the Sciences*](https://www.coursera.org/learn/sciwrite) course at Stanford (2012). ## What This Skill Does Clear writing is not a cosmetic concern in science; it determines whether reviewers and readers can follow your argument. Yet most graduate students receive little formal training in scientific writing, and self-editing for clarity is difficult because you already know what you meant to say. This skill encodes a systematic editorial review process distilled from 30 video lectures on scientific writing. When loaded into an AI tool, it performs five sequential audit passes on your manuscript text: 1. **Clutter extraction.** Flags dead-weight phrases, unnecessary jargon, and filler language, and provides concise replacements. 2. **Voice and verb vitality.** Identifies passive constructions, smothered verbs (nominalizations), and weak verb choices, and suggests active alternatives. 3. **Sentence architecture.** Evaluates sentence length, structure, and flow—flagging run-on sentences, misplaced modifiers, and paragraphs that lack a clear logical progression. 4. **Keyword consistency.** Checks that key technical terms are used consistently throughout the manuscript, catching cases where synonyms introduce ambiguity. 5. **Numerical and citation integrity.** Audits numerical values, units, and in-text citations for internal consistency (e.g., a value reported as 3.2 in the text but 3.4 in a table). The skill supports four review modes: a full-manuscript review that runs all five passes, a single-section review, a targeted review that runs only the pass
话题
暂无话题