Aigc Trace Cleaner

Community

jsChen-biostat

aigc-trace-cleaner

检测并消除论文/报告的 AI 生成痕迹（AIGC 味、AI 味太重、一眼看出是 AI 写的）：10 大维度逐段扫描（词汇/句式/段落/逻辑/人味/数据表述/图表/结构/引用/整体 AI 感），输出带定位与修改示例的诊断报告，并按需改写降低痕迹；另含数学建模竞赛特有的 AI 破绽检查（只仿真不建模、NP 难硬算、图表一键同质化、文献幻觉、逻辑断层、无闭环迭代），以及**2026 国赛 AI 合规全自动自查**（6 大维度 26 条红线：AI 新规合规、幻觉与数据真实性、建模逻辑与题意适配、结果合理性、官方写作规范、排版与页数，配套《AI 合规自查报告》）。当用户要求 AI 痕迹检测、降 AI 味、AIGC 率太高、论文被判定 AI 生成、查重系统提示 AI 率超标、把论文改得更像人写的、或要求国赛 AI 合规自查/AI 红线检测/按自查表核查论文时使用。

Overview

Publisher	jsChen-biostat
Repository	`math-modelCN`
Skill name	`aigc-trace-cleaner`
Stars	59
Forks	0
Bundled files	8
License	MIT
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
8 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by jsChen-biostat on GitHub. Read the source before you install it.

Installation

Install the Aigc Trace Cleaner AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/jsChen-biostat/math-modelCN/tree/main/aigc-trace-cleaner
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/jsChen-biostat/math-modelCN.git /tmp/math-modelCN
mkdir -p .claude/skills
cp -r /tmp/math-modelCN/aigc-trace-cleaner .claude/skills/aigc-trace-cleaner

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Aigc Trace Cleaner in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Aigc Trace Cleaner on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Aigc Trace Cleaner is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

AIGC 痕迹检测与消除（aigc-trace-cleaner）

把"AI 生成痕迹过重、一眼被看出"变成可检测、可定位、可修的问题。方法论取自五份实战文档：10 维检测体系（对齐「终极提示词 V5.0」）+ 数模竞赛 AI 破绽清单 + AI 写作 8 大痕迹 + 2026 国赛 AI 合规自查表（6 维 26 条）。

模式判断（先定模式，再动手）

诊断模式（默认）：只出检测报告与修改建议，不改写正文。适用于"帮我看看这篇论文 AI 味重不重"。
改写模式：用户明确要求"顺手改掉/降到 X% 以下"时才改写；改写前必须先出诊断报告（先诊断后开药）。
合规自查模式（国赛 AI 红线）：用户要"国赛 AI 合规自查/AI 红线检测/按自查表核查"时启用——按 6 维 26 条逐条给出「合格/存在问题/待核实」，出《AI 合规自查报告》；细则见 references/national-competition-compliance.md，报告格式见 references/compliance-report-template.md。
质检模式：只要机械指标与自检清单（跑 scripts/scan.mjs 出统计 + 按自检表核对），适用于快速抽查或改完后复检。

用户没说清楚时用诊断模式，并在报告末尾问是否进入改写模式。合规自查模式需要四件套输入：参赛论文 + 比赛题目 + 附件（如有）+ 自查表；缺项按要求标"待核实（缺 X）"，不得凭空判合格。两种报告可同时交付（诊断报告 + 合规自查报告）；用户明确要"一份完整报告"时才合并编排（见 compliance-report-template.md 末节）。

工作流程（六步，诊断模式）

取全文：要求用户提供完整论文（或至少摘要+正文+结论；图表要含标题与说明）。docx 先转文本（可用 scripts/scan.mjs --docx 或先解压取 word/document.xml）。
机械统计（省 token 关键）：先跑 scripts/scan.mjs，拿到客观数字：段落长度分布与变异系数、句长分布、AI 高频词频次与阈值命中、的字密度、过渡词密度、模板化动词数、空洞形容词连用、小数位异常、设问/破折号等"人味指标"计数。不要靠肉眼数。
逐段扫描定位：从摘要到结论逐段过，结合机械统计标出可疑位置（章节+段落+句），记原文摘录。
10 维深度归因：对每个问题判定维度、严重度（🔴/🟡/🟢）、成因（AI 为什么会这样写、人类通常怎么写）——细则见 references/detection-dimensions.md。
给可执行改法与示例：每条至少一个"原文 → 修改后"对照；禁止"建议优化表达"这类空话——手法见 references/rewrite-rules.md，词表见 references/ai-word-library.md。
出报告：按五部分结构（总体评估 / 详细问题清单 / 维度汇总 / 优先级路线图 / 附录）输出，模板见 references/report-template.md。

数模/竞赛论文：第 3 步后加读 references/math-modeling-pitfalls.md，额外核查 6 大坑（机理缺失、硬算不简化、图表同质化、文献幻觉、逻辑断层、无迭代痕迹）。 合规自查模式：机械统计后加读 references/national-competition-compliance.md，按 6 维 26 条逐条判定（合格/存在问题/待核实），出《AI 合规自查报告》。

交付契约

报告必须含：论文基本信息、A–E 级 AI 生成概率 + 百分比、10 维得分、严重/中等/轻微问题计数；
每条问题必须含：编号（P-001…）、精确位置（章节+段+句）、原文摘录、所属维度、问题类型、严重度、深度分析、修改建议、修改示例（原文→修改后）；
问题清单按论文顺序排列（不是按维度或严重度）；
报告格式：默认 Markdown；用户要 HTML 时按 references/report-template.md 的样式与卡片结构出（可直接浏览器打开）；
合规自查报告（新模式）：五段结构——自查结果汇总表 → 逐条核查明细（26 条，每条含结论/位置/原文证据/判定理由/整改建议）→ 问题分级汇总（致命扣分/一般缺陷/轻微瑕疵）→ 优先级整改清单（按扣分权重排序）→ 官方规范适配度与风险预警；模板见 references/compliance-report-template.md；用户要 Word/PDF/HTML 时按其"输出与交付要求"执行；
改写模式额外交付：改写后的全文 + 改动清单（改了哪些句、为什么），并复跑一次扫描做前后对比。

硬规则（红线）

只改表达、不改内容：数据、公式、结论、逻辑关系一律不动；为了降 AI 味改数字=学术不端。
人味适量：口语化/个人化表达每 1000 字 1–2 处，过量反而显假；全文风格保持一致。
不改坏逻辑：改写必须保持原论证链条，禁止为求变化引入错误因果。
文献不得编造：发现疑似幻觉引用（查不到、作者年份对不上、与论点无关）必须列为🔴并提示核验，不许"顺手美化"。
宁多勿少：诊断阶段宁可多标一个轻微问题，不可漏一个严重问题；但严重度判定必须给出理由。
禁止一次性大替换：批量替换前逐条确认，避免把正确用法一起替换（禁忌见 references/rewrite-rules.md）。
不承诺"过检测"：只承诺降低可识别痕迹；不得声称能保证通过任何 AIGC 检测系统。
合规自查不得和稀泥：每条只允许"合格/存在问题/待核实"；没有证据的一律"待核实"，禁止把未核验项写成合格；文本层查不到的（页数、单页不跨页、三线表、配色、匿名信息在页眉页脚）必须列入"需人工核验"。
官方硬指标优先：合规模式下的摘要字数（800–1000）、正文页数（20–32，最优 25–30）、假设条数（4–6）、三大检验（误差/灵敏度/稳健性）、匿名合规属硬扣分项，必须在报告中逐项给"当前值 → 官方要求 → 差距"。

渐进加载指引（省 token）

触发后先只读本文件；按需再读：
- 判定问题维度/阈值/自检 → references/detection-dimensions.md
- 查词、找人类替代表达 → references/ai-word-library.md
- 具体改写手法与禁忌 → references/rewrite-rules.md
- 出 HTML 痕迹报告 → references/report-template.md
- 数学建模/竞赛论文加检 → references/math-modeling-pitfalls.md
- 国赛 AI 合规自查（26 条红线与硬指标） → references/national-competition-compliance.md
- 出《AI 合规自查报告》/合并版报告 → references/compliance-report-template.md
机械统计一律交给 scripts/scan.mjs（零依赖 Node，默认输出含"10 维痕迹指标 + 国赛合规机械核查"两段），不要用模型逐字数；
不要一次性加载全部 references。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Aigc Trace Cleaner AI skill do?

Why use Aigc Trace Cleaner on TypingMind?

Because you install it once and use it with any model. Aigc Trace Cleaner is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Aigc Trace Cleaner in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/jsChen-biostat/math-modelCN/tree/main/aigc-trace-cleaner. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Aigc Trace Cleaner?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Aigc Trace Cleaner?

As many as you like. As long as a model supports skills, you can use Aigc Trace Cleaner with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Aigc Trace Cleaner AI skill free?

Yes. It is published on GitHub by jsChen-biostat under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Cloud Image Bake

openclaw

OrganizationPopular

Bake, select, prove, and safely retire a Cloud Worker image with crabbox and config one-liners.

Acp Router

Route plain-language requests for Claude Code, Cursor, Copilot, OpenClaw ACP, OpenCode, Gemini CLI, Qwen, Kiro, Kimi, iFlow, Factory Droid, Kilocode, or explicit ACP harness work into either OpenClaw…

Diffs

Use the diffs tool to produce real, shareable diffs (viewer URL, file artifact, or both) instead of manual edit summaries.

Feishu Doc

Feishu document read/write workflows. Activate when the user mentions Feishu docs, cloud docs, or docx links.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

AIGC 痕迹检测与消除（aigc-trace-cleaner）

模式判断（先定模式，再动手）

工作流程（六步，诊断模式）

交付契约

硬规则（红线）

渐进加载指引（省 token）

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now