论文写作

Yichen-ZJU2026-10-27查看 1175复制 957

将研究发现转化为带章节、公式和引用的精炼论文式草稿。当用户要求写论文、起草报告、撰写发现,或根据已收集的研究产出技术文档时使用。

Download (.md)

📘如何使用这个技能

  1. 1
    复制技能内容·点击上方「Copy SKILL.md」复制全部内容,或 Download 下载到本地
  2. 2
    打开你的 AI 客户端·支持 Claude Code、Trae、Cursor、ChatGPT、DeepSeek、Kimi 等任意 AI 助手
  3. 3
    粘贴为系统提示/知识文件·Claude Code:放入 .claude/skills/ 目录;普通对话:作为首条消息发送并说「请严格按这份技能文档执行」
  4. 4
    直接提出你的需求·AI 会按技能定义的工作流、格式和约束来处理,无需再解释背景

💡 提示:技能生效后,每次对话只需一句需求,AI 会按 SKILL.md 的流程执行。

SKILL.md3064 chars
---
name: paper-writing
description: Turn research findings into a polished paper-style draft with sections, equations, and citations. Use when the user asks to write a paper, draft a report, write up findings, or produce a technical document from collected research.
argument-hint: <topic>
---

# Paper Writing

Write a paper-style draft for the user's topic.

Derive a short slug from the topic (lowercase, hyphens, no filler words, ≤5 words). Use this slug for all files in this run.

## Tools (Claude Code)

- Web search: `WebSearch`. Fetch pages: `WebFetch`.
- Academic papers: the `alphaxiv` MCP tools — `discover_papers` (search), `get_paper_content` (read), `answer_pdf_queries` (Q&A on a paper's PDF), `read_files_from_github_repository` (paper code). If these tools are not visible, fall back to WebSearch/WebFetch on arxiv.org and record the degradation.
- Subagents: the `Agent` tool with `subagent_type` `writer` / `verifier`. Real output goes to files.

## Requirements

- **Whole-draft or core-structure rewrites**: first read — or establish via `paper-narrative` — the shared narrative plan at `outputs/.plans/<slug>.md` (mainline, claim–evidence map, section order). Write to that plan; do not invent a parallel story. **Language-only polish stays narrow**: it must not expand into a full research pipeline (no new experiments, no re-deriving the storyline) unless the user asks.
- Before writing, outline the draft structure: proposed title, sections, key claims to make, source material to draw from, and a verification log for the critical claims, figures, and calculations. Write the outline to `outputs/.plans/<slug>.md`. Briefly summarize the outline to the user and continue immediately. Do not ask for confirmation or wait for a proceed response unless the user explicitly requested outline review.
- Use the `writer` subagent when the draft should be produced from already-collected notes, then use the `verifier` subagent to add inline citations and verify sources.
- Include at minimum: title, abstract, problem statement, related work, method or synthesis, evidence or experiments, limitations, conclusion.
- Use clean Markdown with LaTeX where equations materially help.
- Follow the provenance rules for all results, figures, charts, images, tables, benchmarks, and quantitative comparisons. If evidence is missing, leave a placeholder or proposed experimental plan instead of claiming an outcome.
- Generate charts only when real source-backed quantitative data supports them (e.g. matplotlib via Bash); otherwise write a chart specification or table. Use Mermaid for architectures and pipelines only when the structure is supported by sources. Every figure, chart spec, or table needs provenance.
- Before delivery, sweep the draft for any claim that sounds stronger than its support. Mark tentative results as tentative and remove unsupported numerics instead of letting the verifier discover them later.
- Save exactly one draft to `papers/<slug>.md`.
- End with a `Sources` appendix with direct URLs for all primary references.
来源:skillsdirectory

评论0

登录后参与评论

还没有评论,来抢沙发吧