paper-outline
Skill0 starsChanged 2 months agoMITView on GitHub
为 MCM、国赛或期刊论文设计章节结构和 Claim-Evidence-Link 故事线。
2 other projects use a version of this.
Adopted by
Most recently taken up by
- kreid415/claude-science-skillstheir own version, first seen Oct 3, 2026
- DanaResearchGroup/agent-skillstheir own version, first seen Aug 26, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- DanaResearchGroup/agent-skills★ 1same name
- kreid415/claude-science-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository34 other files it gives its agents.
- ahpSkill
- arxiv-literature-synthesisSkill
- ask-mathxSkill
- bp-neural-networkSkill
- chart-decisionSkill
- citation-bibliographySkill
- clustering-classificationSkill
- conjecture-formulationSkill
- counterexample-searchSkill
- diagram-schematicSkill
- differential-equationSkill
- evaluation-entropy-topsisSkill
- figure-table-generationSkill
- genetic-algorithmSkill
- graph-networkSkill
- gray-predictionSkill
- ingest-inputsSkill
- latex-typesettingSkill
- literature-reviewSkill
- monte-carlo-simulationSkill
- numerical-verificationSkill
- optimization-lp-milpSkill
- plotly-interactiveSkill
- polish-proofreadSkill
- proof-assistantSkill
- publication-figureSkill
- regression-familySkill
- reproducibility-checklistSkill
- research-evidenceSkill
- scientific-plottingSkill
- setup-mathxSkill
- statistical-plotSkill
- symbolic-computationSkill
- time-series-arimaSkill
The file
Files
论文大纲与故事线
先读仓库根目录的 CONTEXT.md(术语/单位/venue 默认),全文用它的词。
人触发的生产型技能,只做一件事:建模结果(或白纸)变成逐节大纲,每节有主张、有证据、有连接句。不填 LaTeX 模板(那是 latex-typesetting),不润色文字(那是 polish-proofread)。
Operating Posture
用户键入时运行,论文开头或草稿逻辑崩了时。产品是合作者照着写不用问问题的大纲。标准是每节 Claim-Evidence-Link:抽掉某节主张大纲 visibly 断掉——装饰节在这里砍,不留给审稿人。
两种失败模式,第一种更糟:
- 无主张的标题。“3.2 模型求解”无断言、无证据指针、无字数预算。纯标题大纲把思考外包给动笔时,到那时已经晚了。
- 故事线倒讲。 方法跑在它回答的问题前面,结论超出证据,局限藏进未来工作。大纲是逻辑受压测试的地方,不是装饰的地方。
无主张不给节。无主张,无节。
Hard Rules
- venue 和题型先行。 MCM / 国赛 / journal——结构、篇幅、证据 bar 跟 venue。哪都适配的大纲哪都不适配。
- 每节一主张,写成句子。 不是题目(“ARIMA 结果”)而是断言(“SARIMA(1,1,1)(1,1,1)₁₂ 样本外 RMSE 赢季节朴素 18%”)。写不出的主张暴露缺失的结果——打回方法技能。
- 每主张一证据指针。 图/表编号或待跑的计算,点名。“证据 TBD”的主张是标记过的洞,计数列单——永不沉默。
- 节间连接句。 一句一个:本节结论怎么 feed 下节前提。断掉的连接是审稿人写“disconnected”的地方。
- 每节字数预算。 总量 venue 封顶;分配跟论证权重(结果和方法拿最多)。40% 背景的大纲是穿论文衣服的综述。
Build Sequence
1. 定 venue、题型、问题
- venue、页数/字数上限、中心问题一句话。草稿或白纸——有草稿先诊断逻辑故障(哪个主张无证据?哪段连接断了?)。
- venue 是国赛,走专用页序(2026 修订稿),先把格子定死再谈故事线:
| 页 | 内容 | 硬约束 |
| p1 | 承诺书 | 纸质版有,电子版没有 |
| p2 | 编号专用页 | 纸质版有,电子版没有 |
| p3 | 摘要专用页 | 标题+关键词,无需英文,原则上 ≤1 页;页码从此页起,页脚中部,阿拉伯数字从 1 连续编号 |
| p4 起 | 正文 | 不要目录,不超过 30 页 |
| 正文后 | 附录 | 页数不限,打印装订一起交;含支撑材料文件列表 + 全部可运行源程序(无程序则声明“本论文没有用到程序”) |
- 国赛字数预算按页分配:摘要 1 页封顶,正文 30 页封顶;匿名是页序的一部分——摘要/正文/附录任何地方不许出现身份、学校、赛区信息(第六条)。
2. 三行故事线
- 开局(缺口 + 问题)→ construction(方法 + 证据)→ 收获(答案 + 局限)。收获还说不出,结果没 ready——点名先跑哪个方法技能。
3. 展开到节——gate
- 每节:主张句、证据指针、连接句、字数预算。用下面模板。
- gate:主张自上而下读——推出收获吗?证据列读——TBD 都是带主技能的 listed 洞。洞可以有,隐藏的洞不行。
<outline-template>Section: [编号 + 标题]
Claim: [一句断言]
Evidence: [图/表/计算引用,或 TBD → 主技能]
Link: [本节结论如何 feed 下节]
Budget: [字数]
</outline-template>大纲模板见
assets/templates/outline-<venue>.md(outline-guoshai2026/outline-mcm/outline-journal),按venue_template取用,直接填四行,不要重起格式。
4. 迭代检测 + 移交
- 是否已存在草稿 gate:检查输出目录是否已有
outline_v*.md。无则create新稿;有则走iterate分支——调assets/scripts/versioned_write.py --in 新稿 --out-dir <out> --base-name outline --message "修订说明",产出带时间戳版本 +revisions.md追加记录,不覆盖旧稿(对标 patent 的merger+ 时间戳 + 修订对话记录)。 - 按序点名下个技能(方法跑 →
figure-table-generation→latex-typesetting→citation-bibliography→polish-proofread→reproducibility-checklist)。大纲交付的标准:别人照它能独立执行整条链。
Never Ship
| 禁忌 | 替代 |
| 题目式标题 | 主张句 |
| 证据 TBD 不列 | 列洞、计数、定主 |
| 无连接句 | 每节一条 |
| 无字数预算 | venue 封顶分配 |
| 收获超出证据 | 自上而下压测蕴含关系 |
| 国赛大纲带目录 | 删目录(第四条) |
| 国赛摘要超一页 | 压缩到一页(第三条) |
| 页序出现承诺书/编号页(电子版) | 只放纸质版(第十条) |
Output
- 大纲——上模板逐节。
- 洞清单——TBD 证据带主技能。
- 移交链——下个技能按序。
Tone
立场鲜明、废话少。当正确答案是“第 4 节主张要的结果你没有——先跑模型”就打回。草稿逻辑在 2、3 节间断了,点名断裂,不要改编号绕过去。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

