---
id: gh-auto-empirical-research-skills
name: "Auto-Empirical-Research-Skills"
url: https://skills.yangsir.net/skill/gh-auto-empirical-research-skills
author: brycewang-stanford
domain: science
tags: ["empirical-research", "causal-inference", "econometrics", "academic-writing", "replication"]
install_count: 4729
rating: 4.40 (189 reviews)
github: https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills
---

# Auto-Empirical-Research-Skills

> 把实证研究需求路由到最匹配技能，覆盖因果推断、计量经济学、复现、数据获取、论文写作、同行评审、引用核查与去 AIGC 编辑。适合社会科学研究者从原始数据一路推进到规范可复现的 Word 论文，避免一次性加载整个仓库。

**Stats**: 4,729 installs · 4.4/5 (189 reviews)

## Before / After 对比

### 实证论文全流程技能路由

**Before**:

使用前，研究者要在庞大仓库中手动翻找因果推断、数据清洗、复现与写作技能，容易选错技能、重复读文档，流程搭建耗时且不可控。

**After**:

使用后，按任务阶段自动定位单个 SKILL.md，直接获得方法选择、脚本模板与写作规范，从原始数据到 Word 论文的路径更清晰、更可复现。

| Metric | Before | After | Change |
|---|---|---|---|
| 技能选择耗时 | 30分钟 | 5分钟 | -83% |

## Readme

---
name: auto-empirical-research-skills
description: Route empirical-research requests through the Auto-Empirical Research Skills catalog when this whole repository is installed as one skill in Codex, CodeBuddy, Claude Code, or another IDE. Use to choose and load the right vendored AERS skill for causal inference, econometrics, replication, data acquisition, manuscript writing, peer review and referee responses, citation checking, de-AIGC editing, or full empirical-paper workflows, or an end-to-end run from raw data all the way to a finished Word (.docx) manuscript, without reading the entire repository at once.
license: CC-BY-SA-4.0
---

# Auto-Empirical Research Skills Router

Use this root skill when the full AERS repository has been installed as a single skill folder. Treat it as a router and catalog, not as a request to load every vendored `SKILL.md`.

The catalog holds **1,107 skills across 77 vendored collections**. Never read them all — route to one, then load only that skill's `SKILL.md`.

## Workflow

1. Classify the user's empirical-research task by **stage**, then load the single best-matching skill:
   - Full pipeline or orchestration: start with `skills/69-Paper-WorkFlow/` or the `skills/00*` flagship analysis skills — `skills/00-Full-empirical-analysis-skill_StatsPAI/` (StatsPAI), `skills/00.1-Full-empirical-analysis-skill_Python/` (Python), `skills/00.2-Full-empirical-analysis-skill_Stata/` (Stata), `skills/00.3-Full-empirical-analysis-skill_R/` (R). Note the StatsPAI flagship has no dot in its prefix, so a `skills/00.*` glob misses it.
   - Causal inference and econometrics: pick by method from the table below, or search `catalog/skills.json` / `docs/TAXONOMY.md`.
   - AER or top economics journal work: start with `skills/50-brycewang-aer-skills/`.
   - Replication, citation, or peer review: use `docs/SKILL_CATALOG.md` and `docs/GOLDEN_WORKFLOWS.md` to choose a focused skill.
   - Academic de-AIGC (English or Chinese) or academic rewriting: start with `skills/48-de-AIGC-skills/` or nearby writing skills in the catalog.
2. Read only the selected child skill's `SKILL.md`, then follow its progressive-disclosure instructions for `references/`, `scripts/`, `assets/`, or templates.
3. If no child skill clearly matches, run the ranked catalog search first — it takes English or Chinese task text:

     ```bash
     python3 scripts/find-skill.py "staggered difference-in-differences event study"
     python3 scripts/find-skill.py "降低中文论文 AIGC 率" -k 5
     ```

   It ranks by relevance, then by the routing tier in `catalog/curation.json`: first-party `core` skills first, a single preferred copy for each duplicated name (the other copies are folded away), and natural-science / CS guides (`out-of-domain`) last unless the query names them. Take the top result unless its description clearly misfits, then read only that `SKILL.md`. Its accuracy is measured by `python3 scripts/check-routing.py` against `evals/routing-cases.json`.

   If you need raw data instead, `catalog/skills.json` has `path`, `name`, `description`, `line_count`, and a globally-unique `qualified_name`; `catalog/skills-enriched.json` adds `tier`, `tags`, `quality_score`, `license`, and `commercial_use`. Avoid broad recursive reads of `skills/`.
   - Both catalog JSON files are large (roughly 1 MB / 20k lines each) — query them instead of reading them whole. Example:

     ```bash
     python3 -c "import json; [print(s['qualified_name'], '->', s['path']) for s in json.load(open('catalog/skills.json'))['skills'] if 'synthetic control' in (s['name'] + ' ' + s['description']).lower()]"
     ```

     A plain `grep -in "synthetic control" catalog/skills.json` works too when a rough match is enough.
4. For installation help, use `docs/INSTALL.md` for Codex-style copy installs and `INSTALL.md` for Claude Code marketplace/plugin installs.
5. If editing this repository, keep parent and nested repos separate. In particular, inspect `git status` inside `skills/69-Paper-WorkFlow/` (a git submodule) before touching it.

## Method → where to start

Match the user's identification strategy or task to a starting collection, then confirm against `catalog/skills.json`.

This table is a shortcut to the most common starting points, **not a complete index** — it names fewer than half of the vendored collections, and the rest are reachable only through `catalog/skills.json`. A task missing from this table is not a task without a skill: fall through to step 3 and search the catalog before concluding nothing matches.

| Task / method | Start here |
|---|---|
| Full paper pipeline (orchestrator) | `skills/69-Paper-WorkFlow/` |
| **Data → full Word `.docx` manuscript** (one run: analysis + writing + assembled deliverable) | `skills/69-Paper-WorkFlow/` — pick `manuscript.format = markdown` at its Stage 0 when the deliverable is Word; Stage 9 assembles `09_submission/main.docx` (body + tables + figures + references) and gates it |
| Markdown / LaTeX → `.docx` conversion only (no analysis) | `skills/67-econfin-workflow-toolkit/md-to-docx/`, `skills/08-ndpvt-web-latex-document-skill/` |
| Agent-native causal analysis (one call runs DiD / RD / IV / SCM / DML with automatic robustness gates) | `skills/00-Full-empirical-analysis-skill_StatsPAI/` |
| DiD / staggered DiD / event study | `skills/50-brycewang-aer-skills/`, `skills/10-Jill0099-causal-inference-mixtape/`, `skills/13-scunning1975-MixtapeTools/` |
| Instrumental variables (IV) | `skills/50-brycewang-aer-skills/`, `skills/40-py-econometrics-pyfixest/` |
| Regression discontinuity (RDD) | `skills/50-brycewang-aer-skills/`, `skills/10-Jill0099-causal-inference-mixtape/` |
| Synthetic control (SCM) | `skills/50-brycewang-aer-skills/`, `skills/13-scunning1975-MixtapeTools/` |
| Panel fixed effects | `skills/40-py-econometrics-pyfixest/`, `skills/39-vincentarelbundock-marginaleffects/` |
| Specification search / multiverse / p-hacking audit, honest p-values, p-curve, pre-registration | `skills/73-brycewang-p-hacking-skills/` (start at `skills/00-phack-router/`; teaching & audit only) |
| Matching / propensity scores | `skills/10-Jill0099-causal-inference-mixtape/`, `skills/11-James-Traina-compound-science/` |
| Structural estimation | `skills/11-James-Traina-compound-science/`, `skills/14-luischanci-claude-code-research-starter/` |
| Time series / forecasting | `skills/17-DAAF-Contribution-Community-daaf/`, `skills/43-wentorai-research-plugins/` |
| Text as data / NLP | `skills/43-wentorai-research-plugins/` |
| Spatial / GIS analysis | `skills/17-DAAF-Contribution-Community-daaf/`, `skills/43-wentorai-research-plugins/` |
| Experiments / RCT design | `skills/11-James-Traina-compound-science/`, `skills/25-HosungYou-Diverga/` |
| Survey / questionnaire design | `skills/43-wentorai-research-plugins/`, `skills/25-HosungYou-Diverga/` |
| DML / CATE / causal forests | `skills/00.1-Full-empirical-analysis-skill_Python/`, `skills/63-tondevrel-scientific-agent-skills/` |
| Bayesian modeling | `skills/23-Learning-Bayesian-Statistics-baygent-skills/`, `skills/51-pymc-labs-CausalPy/` |
| Python analysis (full pipeline) | `skills/00.1-Full-empirical-analysis-skill_Python/`, `skills/40-py-econometrics-pyfixest/` |
| Stata analysis | `skills/00.2-Full-empirical-analysis-skill_Stata/`, `skills/32-dylantmoore-stata-skill/`, `skills/64-tmonk-mcp-stata/` |
| R analysis | `skills/00.3-Full-empirical-analysis-skill_R/`, `skills/55-ab604-claude-code-r-skills/` |
| Game theory / theory papers | `skills/65-game-theory-paper-writer/` |
| Qualitative / thematic analysis | `skills/53-keemanxp-thematic-analysis-skill/` |
| Data acquisition (Kaggle, SEC filings, open data) | `skills/72-kaggle-research/`, `skills/57-dgunning-edgartools/`, `skills/59-shiquda-openalex-skill/` |
| Literature review | `skills/36-taoyunudt-literature-review-skill/`, `skills/52-keemanxp-slr-prisma/`, `skills/59-shiquda-openalex-skill/` |
| Lit-review tool selection / PDF→Markdown / cited Q&A over PDFs / PRISMA screening runners | `skills/71-brycewang-lit-review-agent-tools/` |
| Citation checking | `skills/62-PHY041-claude-skill-citation-checker/` |
| Manuscript writing / proofreading | `skills/04-K-Dense-AI-claude-scientific-writer/`, `skills/38-peternka-academic-proofreader/` |
| Peer review / referee reports / referee responses | `skills/21-claesbackman-AI-research-feedback/`, `skills/12-pedrohcgs-claude-code-my-workflow/`, `skills/67-econfin-workflow-toolkit/` |
| LaTeX / Quarto compilation, slides | `skills/08-ndpvt-web-latex-document-skill/`, `skills/60-regisely-superpapers/`, `skills/12-pedrohcgs-claude-code-my-workflow/` |
| De-AIGC / humanize | `skills/48-de-AIGC-skills/`, `skills/45-stephenturner-skill-deslop/`, `skills/47-conorbronsdon-avoid-ai-writing/` |
| Chinese SSCI/CSSCI journal polishing | `skills/70-ssci-polish/`, `skills/49-voidborne-d-humanize-chinese/` |
| Replication | `skills/28-maxwell2732-paper-replicate-agent-demo/`, `skills/29-quarcs-lab-project20XXy/` |
| Open science / reproducibility | `skills/54-scdenney-open-science-skills/`, `skills/29-quarcs-lab-project20XXy/` |
| Grant proposals / funding | `skills/42-wanshuiyin-ARIS/`, `skills/43-wentorai-research-plugins/` |
| Conference posters / post-acceptance | `skills/42-wanshuiyin-ARIS/`, `skills/33-Galaxy-Dawn-claude-scholar/` |

## Full-pipeline trigger

If the user is asking for a complete empirical paper from idea to submission, route to `skills/69-Paper-WorkFlow/`. The orchestrator loads the right skill at the right stage and stops for human decisions at the two hard gates (Method Gate after Stage 3, Draft Quality Gate after Stage 7).

Trigger phrases (any one is enough to dispatch to the orchestrator):

- `/paper-workflow`
- "帮我写一篇实证论文"
- "从选题到投稿"
- "end-to-end empirical paper"
- "完整复现"
- "from proposal to submission"
- "从数据到 docx 论文全文" / "一条龙" / "出一份 Word 版论文"
- "raw data to a finished Word manuscript"

The orchestrator is **not** the right entry point for a single-task ask (e.g. "fit a DiD", "recode this variable", "write a referee report") — those are listed in the Method → where to start table above.

## Coverage Notes

- `skills/69-Paper-WorkFlow/` is a **git submodule**. If its folder is empty, the copy or clone skipped submodules (`git submodule update --init` fixes a clone); fall back to the `skills/00*` flagship pipeline skills, which are vendored directly. Those end at publication-ready tables and figures plus a Step 8.5 handoff contract (`exhibits_index.md` + `results_summary.json`); pair them with a writing skill for the manuscript itself, since assembling and gating the full `.docx` lives in the orchestrator.
- The vendored ARIS collection (`skills/42-wanshuiyin-ARIS/`) also ships its skill set as OpenAI Codex CLI runtime ports (`skills-codex*` subtrees). Those stay on disk but are excluded from `catalog/skills.json` (see `scripts/skill_discovery.py`) — route Claude agents to the primary `skills/` tree only.

## Install Notes

- Whole-repo imports are supported by this root `SKILL.md` as a lightweight compatibility entry point.
- Individual skill installs are still preferred when a runtime expects one folder per skill. Copy the folder that directly contains the target `SKILL.md`.
- Do not copy the repository root into a runtime and expect every child skill to become individually registered unless that runtime explicitly supports recursive skill discovery.
- **Do not flat-install the whole catalog** (e.g. symlinking every child folder into `~/.claude/skills`). Every registered skill's description is loaded at session start — about 64k tokens for all 1,107 here — and runtimes truncate long skill listings, so matching gets *worse*, not better. Use a plugin, this router, or a handful of copied skills.
- **Name collisions:** the catalog contains 47 bare `name`s shared across collections (e.g. `data-analysis`, `lit-review`, `proofread`); `catalog/curation.json` names one preferred copy of each, which `scripts/find-skill.py` returns. When a runtime registers skills by flat name, install one collection at a time, or disambiguate with the globally-unique `qualified_name` field in `catalog/skills.json` (`<collection>::<name>`, e.g. `12-pedrohcgs-claude-code-my-workflow::data-analysis`), or the full `skills/<collection>/.../SKILL.md` path.

## Key Files

- `catalog/skills.json`: machine-readable list of vendored skills.
- `catalog/skills-enriched.json`: same list plus `tier`, `tags`, `quality_score`, `license`, and `commercial_use` for filtering.
- `catalog/curation.json`: hand-curated routing tiers (core collections, the preferred copy of each duplicated name, out-of-domain prefixes). Ranking only — nothing is removed.
- `scripts/find-skill.py`: ranked search over the catalog (the router's search step).
- `docs/SKILL_CATALOG.md`: human-readable skill index.
- `docs/TAXONOMY.md`: task and method taxonomy.
- `docs/GOLDEN_WORKFLOWS.md`: ready-to-use empirical-research prompts.
- `docs/INSTALL.md`: runtime installation guidance for single-skill and whole-repo use.
- `docs/CONTENT_ZH.md` and `README-zh-CN.md`: Chinese-language collection index and entry point. Prefer these when the user is working in Chinese — several collections (de-AIGC, SSCI/CSSCI polishing, Chinese academic writing) are documented there in more detail than in the English docs.

--- README.md ---

[![Awesome](https://awesome.re/badge.svg)](https://awesome.re)
[![GitHub stars](https://img.shields.io/github/stars/brycewang-stanford/Auto-Empirical-Research-Skills?style=social)](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills)
[![License: CC BY-SA 4.0](https://img.shields.io/badge/License-CC%20BY--SA%204.0-lightgrey.svg)](https://creativecommons.org/licenses/by-sa/4.0/)
[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](CONTRIBUTING.md)
[![Validate catalog](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/actions/workflows/validate-catalog.yml/badge.svg)](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/actions/workflows/validate-catalog.yml)
[![OpenSSF Scorecard](https://api.scorecard.dev/projects/github.com/brycewang-stanford/Auto-Empirical-Research-Skills/badge)](https://scorecard.dev/viewer/?uri=github.com/brycewang-stanford/Auto-Empirical-Research-Skills)
[![Security audit: baseline 52/52 CLEAN](https://img.shields.io/badge/security%20audit-baseline%2052%2F52%20CLEAN-brightgreen)](SECURITY-SCAN-REPORT.md)
[![Rigor coverage](https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2Fbrycewang-stanford%2FAuto-Empirical-Research-Skills%2Fmain%2Fdocs%2Fbadges%2Frigor-coverage.json)](docs/RIGOR_COVERAGE.md)
[![Powered by StatsPAI](https://img.shields.io/badge/powered%20by-StatsPAI-orange)](https://github.com/brycewang-stanford/StatsPAI)

# Auto-Empirical Research Skills (AERS)

> ## 🔥 力荐 · [StatsPAI](https://github.com/brycewang-stanford/StatsPAI) — Stata 与 R 的 Python 平替，为 AI Agent 而生
>
> - 🔁 **Stata / R 平替**：`reghdfe`、`csdid`、`rdrobust`、`synth`、`fixest`、`did` 等常用命令在 Python 中一站式替代，一个 `import statspai as sp` 搞定 DID / RD / IV / SCM / DML / matching
> - 🤖 **面向 Agent 设计**：结构化结果对象 + 内置 MCP Server，Claude / Codex 等智能体可直接调用、诊断、审计与出表
> - ⚡ **运行速度更快**：无需启动 Stata / R 会话，免许可证，纯 Python 流水线端到端提速
> - 🏛️ **同一团队维护**：StatsPAI 与 AERS 由同一团队（Stanford REAP）开发和持续维护，两者深度集成、同步迭代
>
> **[github.com/brycewang-stanford/StatsPAI](https://github.com/brycewang-stanford/StatsPAI)** · `pip install statspai` · MIT 开源 · 欢迎 ⭐ Star

<div align="center">

**🌐 语言: [English](README-en.md) | 简体中文（默认） | [繁體中文](README-zh-TW.md) | [日本語](README-ja.md) | [한국어](README-ko.md)**

<br/>

  <table>
    <tr>
      <td align="center">
        <a href="https://github.com/brycewang-stanford/StatsPAI"><img src="images/statspai-logo.png" alt="StatsPAI" width="360" /></a>
      </td>
      <td width="72"></td>
      <td align="center">
        <img src="images/stanford-reap-logo.png" alt="Stanford REAP - Center on China's Economy & Institutions" width="440" />
      </td>
    </tr>
  </table>

  <br/>

  <strong>Stanford REAP × StatsPAI</strong> · 实证研究 AI 工具的学术工业级产品<br/>
  <sub>由斯坦福实证研究方法论团队打造，覆盖从数据清洗到顶刊投稿的完整工作流</sub>

  <br/>
  <br/>

  <img src="images/aers-readme-cover-cn.png" alt="实证研究智能体技能大全封面图" width="100%" />

  <br/>
</div>

> ### 🚀 New here? Open the **[Skill Search →](docs/search.html)** to filter all 1,107 skills by method, stage, language, and license. The 5-minute tour (`make quickstart`) prints the same picture in your terminal.
>
> ### 🇨🇳 **中文用户从本文件开始**（流水线速览 + 77 行总表），每个合集的完整描述见 [`docs/CONTENT_ZH.md`](docs/CONTENT_ZH.md)。📖 **English readers:** see [`README-en.md`](README-en.md).

---

### 信任面 · Trust surface (rigor stats)

| Rigor lane | Count | Where |
|---|---|---|
| Numeric **benchmark tasks** — gold values recomputed from real data each run | **19** | [`benchmark/`](benchmark/) |
| Behavioral **eval scenarios / rubric items** | **42 / 217** | [`eval-harness/`](eval-harness/) |
| 其中**已证明能区分对错**的场景（pass/fail 双 fixture 自检） | **9**（全部 6 个 critical 场景在内） | [`eval-harness/fixtures/`](eval-harness/fixtures/) |

> Full trust overview: [`docs/TRUST.md`](docs/TRUST.md) · [`docs/RIGOR_COVERAGE.md`](docs/RIGOR_COVERAGE.md)
>
> 🏁 **带上你自己的 agent 来考同一份卷子**：`pip install -e .` 后用 [`aers-score`](aers_score/README.md) 给自己打分，成绩发布在 [`docs/EXTERNAL_SCOREBOARD.md`](docs/EXTERNAL_SCOREBOARD.md)（规则见 [`docs/SCOREBOARD_RULES.md`](docs/SCOREBOARD_RULES.md)）。榜上的数字是**我们用同一套评分器重算**出来的，不是提交者自报的。

---

## ⚡ 安装与使用（30 秒上手）

### 最省事的一招：把 URL 丢给 Agent

**把项目 URL 地址 `https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills` 丢给 Claude Code / Codex，并指定是目录 / 项目 / 全局安装** —— 剩下的让它自己做。例如：

```text
帮我安装 https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills
装到「全局」（~/.claude/skills/），我想在所有项目里都能用
```

把最后一行换成你要的作用域即可：

| 作用域 | 说给 Agent 的话 | 落到哪里 |
|:--|:--|:--|
| **目录**（当前会话临时用） | "只在当前目录用，不要全局安装" | 当前工作目录下的 `.claude/skills/` |
| **项目**（团队共享，可提交进 git） | "装到本项目" | 项目根目录 `.claude/skills/` |
| **全局**（所有项目可用） | "装到全局" | `~/.claude/skills/`（Codex 为 `~/.codex/skills/`） |

### 手动安装（两种，任选其一）

**A. 插件市场（Claude Code v2.1+，推荐，可升级）**

```bash
claude plugin marketplace add brycewang-stanford/Auto-Empirical-Research-Skills
claude plugin install aer-skills@auto-empirical-research-skills                 # 顶刊投稿全流程（9 skills）
claude plugin install empirical-analysis-python@auto-empirical-research-skills  # Python 计量流水线
claude plugin install empirical-analysis-stata@auto-empirical-research-skills   # Stata 计量流水线
claude plugin install empirical-analysis-r@auto-empirical-research-skills       # R + Quarto 流水线
```

**B. 只要某一个 skill —— 直接拷文件夹**

```bash
git clone --recurse-submodules https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git
cd Auto-Empirical-Research-Skills

cp -R skills/00.1-Full-empirical-analysis-skill_Python  .claude/skills/   # 项目级
cp -R skills/00.1-Full-empirical-analysis-skill_Python  ~/.claude/skills/ # 全局
```

拷进去的文件夹必须自带 `SKILL.md`（部分合集的 `SKILL.md` 在下一层，拷那一层）。

### 怎么用

新开一个会话，**直接用自然语言说要做什么**，Agent 会按 `description` 自动挑 skill；说不动就点名方法或 skill：

```text
用面板数据跑一个 Callaway–Sant'Anna 事件研究，并出 HonestDiD 稳健性和期刊级表格
```

> 完整安装说明（Codex / CodeBuddy 整库导入、`--plugin-dir` 单次加载、常见故障排查）见 [`INSTALL.md`](INSTALL.md)。

---

## 中文文档结构

中文内容分两级维护，各司其职：

- **本文件（README.md，GitHub 默认入口）**：banner、badges、信任面、9 阶段流水线速览、77 行合集总表。
- **[`docs/CONTENT_ZH.md`](docs/CONTENT_ZH.md)（扩展正文）**：每个合集的完整描述（`#skill-NN` 锚点）、按用途分组、精确数字、2 分钟验证、三层信任、旗舰流水线详解、贡献与引用。总表行内的 `→` 直接跳到对应锚点。
- **其他语言**：[`README-en.md`](README-en.md) · [`README-zh-TW.md`](README-zh-TW.md) · [`README-ja.md`](README-ja.md) · [`README-ko.md`](README-ko.md)

> [!NOTE]
> **维护规则：** 改合集总表 → 本文件与 CONTENT_ZH.md 的锚点表**两处同步**；改合集详情 / 分组 / 数字 → 只改 [`docs/CONTENT_ZH.md`](docs/CONTENT_ZH.md)。统计数字（合集数 / skill 数）以 [`catalog/skills.json`](catalog/skills.json) 为准，由 `make validate` 的 readme-stats 检查器守护。
>
> **贡献者（Contributors）：** 提交前请在本地跑通完整门禁 `make check`（catalog 校验 + 链接 + 单元测试 + eval-harness + benchmark）。详见 [`CONTRIBUTING.md`](CONTRIBUTING.md)。
>
> **旧版归档：** [`README-zh-CN.md`](README-zh-CN.md) 已弃用，仅作向后兼容的重定向占位。

---

## 🚀 从一个 idea 到一篇论文：社科实证研究 · 端到端流水线（全自动、可介入）

> **AERS 不只是 77 个散装 skill —— 它能陪你走完一篇论文。** 从模糊 idea → 选题精炼 → 文献综述 → 数据获取 → 识别策略 → 估计建模 → 稳健性审计 → 出版级表格 / 图形 → 写作与同行评审 → 降 AIGC → 投稿。**端到端、全自动、每一步都可被人介入**（中间任何一步你都可以接过去手工改方法、补变量、加稳健性，再让流水线自动接上跑）。

### 9 阶段流水线 · 每一步都覆盖到具体 skill

| # | 阶段 | 关键 skills（点合集名进目录，`→` 进完整说明） |
|:--:|:--|:--|
| 1️⃣ | **选题精炼** — Agent 把模糊想法收紧成"可证伪 + 可执行"的研究问题 | · [25 Diverga](skills/25-HosungYou-Diverga/) · [33 claude-scholar](skills/33-Galaxy-Dawn-claude-scholar/) · [05 research-superpower](skills/05-kthorn-research-superpower/) · [11 compound-science](skills/11-James-Traina-compound-science/) |
| 2️⃣ | **文献综述** — 检索 · 筛选 · PRISMA 流程 · 批判性阅读 · 主题分析 | · [36 literature-review-skill](skills/36-taoyunudt-literature-review-skill/) · [24 academic-research-skills](skills/24-Imbad0202-academic-research-skills/) · [59 openalex-skill](skills/59-shiquda-openalex-skill/) · [68 research-productivity-skills](skills/68-research-productivity-skills/) · [53 thematic-analysis](skills/53-keemanxp-thematic-analysis-skill/) |
| 3️⃣ | **数据获取** — 公开数据库 · API · 网页抓取 · 数据清洗 | · [33 claude-scholar](skills/33-Galaxy-Dawn-claude-scholar/) · [68 research-productivity-skills](skills/68-research-productivity-skills/) · [32 stata-skill](skills/32-dylantmoore-stata-skill/) · [57 edgartools](skills/57-dgunning-edgartools/) |
| 4️⃣ | **识别策略** — DiD / RD / IV / SCM / DML / matching 全覆盖 | · ⭐ [00 StatsPAI](skills/00-Full-empirical-analysis-skill_StatsPAI/) 🔥 · [10 causal-inference-mixtape](skills/10-Jill0099-causal-inference-mixtape/) · [13 MixtapeTools](skills/13-scunning1975-MixtapeTools/) · [51 CausalPy](skills/51-pymc-labs-CausalPy/) · [63 scientific-agent-skills](skills/63-tondevrel-scientific-agent-skills/) |
| 5️⃣ | **估计建模** — Python / Stata / R 三栈，900+ 估计器 | · ⭐ [00.1 Full Empirical · Python](skills/00.1-Full-empirical-analysis-skill_Python/) · ⭐ [00.2 Full Empirical · Stata](skills/00.2-Full-empirical-analysis-skill_Stata/) · ⭐ [00.3 Full Empirical · R](skills/00.3-Full-empirical-analysis-skill_R/) · [40 pyfixest](skills/40-py-econometrics-pyfixest/) · [39 marginaleffects](skills/39-vincentarelbundock-marginaleffects/) · [09 awesome-econ-ai](skills/09-meleantonio-awesome-econ-ai-stuff/) |
| 6️⃣ | **稳健性审计** — 复现包检查 · Honest-DiD · R&R 模拟 | · [41 sewage-econometrics-check](skills/41-sticerd-eee-sewage-econometrics-check/) · ⭐ [50 AER-skills](skills/50-brycewang-aer-skills/) · ⭐ [73 p-hacking-skills](skills/73-brycewang-p-hacking-skills/) · [21 AI-research-feedback](skills/21-claesbackman-AI-research-feedback/) |
| 7️⃣ | **表格 & 图形** — 期刊出版级排版 · LaTeX 嵌入 | · ⭐ [00 StatsPAI](skills/00-Full-empirical-analysis-skill_StatsPAI/) · [07 AI-Research-SKILLs](skills/07-Orchestra-Research-AI-Research-SKILLs/) · [33 claude-scholar](skills/33-Galaxy-Dawn-claude-scholar/) · [08 latex-document-skill](skills/08-ndpvt-web-latex-document-skill/) |
| 8️⃣ | **写作 & 同行评审** — LaTeX / Quarto · 仿审稿人 · 校对 | · [06 stats-paper-writing](skills/06-fuhaoda-stats-paper-writing/) · [04 scientific-writer](skills/04-K-Dense-AI-claude-scientific-writer/) · [22 christopherkenny-skills](skills/22-christopherkenny-skills/) · [38 academic-proofreader](skills/38-peternka-academic-proofreader/) · [56 econ-writing-skill](skills/56-hanlulong-econ-writing-skill/) · [16 clo-author](skills/16-hsantanna88-clo-author/) |
| 9️⃣ | **降 AIGC & 去水印 & 投稿** — 知网 / 万方 / Turnitin / 34 类 AI 痕迹模式 / 隐藏字符 · C2PA · docx 元数据清理 | · ⭐ [48 de-AIGC-skills](skills/48-de-AIGC-skills/) 🇨🇳🇬🇧 · [44 humanizer_academic](skills/44-matsuikentaro1-humanizer_academic/) · [45 deslop](skills/45-stephenturner-skill-deslop/) · [46 stop-slop](skills/46-hardikpandya-stop-slop/) · [47 avoid-ai-writing](skills/47-conorbronsdon-avoid-ai-writing/) · [49 humanize-chinese](skills/49-voidborne-d-humanize-chinese/) |

### 🎼 元编排：⭐ [69 Paper-WorkFlow](skills/69-Paper-WorkFlow/) —— 一键串起来

[`Paper-WorkFlow`](skills/69-Paper-WorkFlow/) 是 AERS 的"指挥棒"，它把上面 9 个阶段的 skill 串成 **一条按键即运行的端到端流水线**。
你在 IDE 入口给它一句自然语言：

> *"开一个新论文项目：空气污染与中国劳动力市场，CS 设计 + 省级面板"*

它会自动按顺序调：

1. ⭐ [00 StatsPAI](skills/00-Full-empirical-analysis-skill_StatsPAI/) → `sp.csdid(...)` 给出 CS-DID 估计草案 + 写出估计方程与识别假设
2. [33 claude-scholar](skills/33-Galaxy-Dawn-claude-scholar/) → 抓变量定义 / 数据源候选 / 相关文献
3. ⭐ [00 StatsPAI](skills/00-Full-empirical-analysis-skill_StatsPAI/) → 真跑 `sp.feols(...)` + `sp.honest_did(...)`
4. [41 sewage-econometrics-check](skills/41-sticerd-eee-sewage-econometrics-check/) → 10 项复现包审计 + 稳健性体检
5. ⭐ [00 StatsPAI](skills/00-Full-empirical-analysis-skill_StatsPAI/) + [07 AI-Research-SKILLs](skills/07-Orchestra-Research-AI-Research-SKILLs/) → 出 Table 1–5 + 期刊级图
6. [38 academic-proofreader](skills/38-peternka-academic-proofreader/) → 通读 + §comment 标"审稿人会挑刺的位置"
7. [56 econ-writing-skill](skills/56-hanlulong-econ-writing-skill/) 起草初稿 + ⭐ [48 de-AIGC-skills](skills/48-de-AIGC-skills/) 🇨🇳🇬🇧 + [45 deslop](skills/45-stephenturner-skill-deslop/) 过知网 / Turnitin

**任何阶段你都可以手动介入** —— 上一阶段的产物全部落盘（产物-幂等 pipeline），你接过去改方法、补控制、加稳健性，再让流水线自动接下去跑。这就是"全自动 + 可介入"。

### 🏆 7 个 Stanford REAP × StatsPAI 自研 skill —— 是整个流水线的主干

| ⭐ Skill | 在流水线里的角色 |
|:--|:--|
| [00 StatsPAI](skills/00-Full-empirical-analysis-skill_StatsPAI/) 🔥 | **因果引擎**：900+ 函数，`sp.causal(...)` 一行跑闭环（DID / RD / IV / SCM / DML / matching） |
| [00.1 Full Empirical · Python](skills/00.1-Full-empirical-analysis-skill_Python/) 📘 | 显式 Python 栈（pandas / statsmodels / linearmodels / pyfixest） |
| [00.2 Full Empirical · Stata](skills/00.2-Full-empirical-analysis-skill_Stata/) 📊 | 显式 Stata 栈（reghdfe / ivreg2 / csdid / sdid / rdrobust） |
| [00.3 Full Empirical · R](skills/00.3-Full-empirical-analysis-skill_R/) 📗 | 显式 R 栈（tidyverse / fixest / did / HonestDiD）+ Quarto 渲染 |
| [48 de-AIGC-skills](skills/48-de-AIGC-skills/) 🇨🇳🇬🇧 | 中英双语学术降 AIGC + 去水印层（Turnitin AI / GPTZero / 知网 / 万方 · 隐藏字符 / C2PA / docx 元数据） |
| [50 AER-skills](skills/50-brycewang-aer-skills/) 📕 | Top-5 经济学投稿套件：识别 → 稳健性 → R&R |
| [69 Paper-WorkFlow](skills/69-Paper-WorkFlow/) 🧭 | 元编排器，把上面 9 个阶段串成一键流水线 |

**为什么挑这 7 个？因为它们的行为都被基准钉死了** —— 不是营销口径，是对着已知答案反复跑过验证过的（[17 项数值 benchmark + 37 项行为评测 ↗](docs/CONTENT_ZH.md#你究竟得到什么精确数字)）。

### 看到这里 —— 完整 77 行合集目录

[↴ 直跳到下方 77 行总表（每个合集带 `#skill-NN` 锚点）](#-77-个核心-skills-合集一览00--73编号连续无空缺)。如果你更关心"这些 skill **怎么用**"而不是"有哪些 skill"，看 [📘 中文唯一权威正文](docs/CONTENT_ZH.md) 里的「按用途分组」与「旗舰流水线」两节。

---

## 🧰 77 个核心 Skills 合集一览（`00 → 73`，编号连续无空缺）

> **打开仓库 → 看见整座库。** 全部 **77 个合集 · 1,107 个 skill**，每一个都已 vendor 进本仓库，由 [`catalog/skills.json`](catalog/skills.json) 跟踪。**⭐ = Stanford REAP × StatsPAI 团队自研的 skill**；其余为精选、经安全审计的社区作品。
>
> **主题图例 —** 🚀 全流程与编排器 · 🎯 因果推断与计量经济学 · 📚 文献与研究设计 · ✍️ 写作 / 编辑 / 去 AIGC · 📑 引用 / 复现 / 同行评审 · 🛠️ 数据 / 工具 / 基础设施
>
> **点击【→】** 跳转到 [`docs/CONTENT_ZH.md`](docs/CONTENT_ZH.md) 中该合集的完整描述；**点击合集名** 直接打开其目录。
>
> **🙏 尊重原作者 —** **「来源」列直接链回上游原始仓库**（`owner/repo`）。本仓库里的社区合集都是**上游快照**：请去原仓库点 star、提 issue、看 LICENSE。完整的许可证与来源置信度审计见 [`docs/LICENSE_AUDIT.md`](docs/LICENSE_AUDIT.md)，机器可读版本在 [`catalog/provenance.json`](catalog/provenance.json)。

| # | 合集 | 一句话 | 详情 | 来源 |
|--:|:--|:--|:--|:--|
| ⭐ [00](skills/00-Full-empirical-analysis-skill_StatsPAI/) | **StatsPAI** 🔥 | 因果引擎 · Agent-native Python DSL：`sp.causal(...)` 一行跑闭环（DID/RD/IV/SCM/DML，900+ 函数） | [→](docs/CONTENT_ZH.md#skill-00) | [brycewang-stanford/StatsPAI](https://github.com/brycewang-stanford/StatsPAI) |
| ⭐ [00.1](skills/00.1-Full-empirical-analysis-skill_Python/) | **Full Empirical · Python** 📘 | 显式栈：`pandas` · `statsmodels` · `linearmodels` · `pyfixest` | [→](docs/CONTENT_ZH.md#skill-00-1) | [⭐ 本仓库](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills) |
| ⭐ [00.2](skills/00.2-Full-empirical-analysis-skill_Stata/) | **Full Empirical · Stata** 📊 | `reghdfe` · `ivreg2` · `csdid` · `sdid` · `rdrobust` 复现包 | [→](docs/CONTENT_ZH.md#skill-00-2) | [⭐ 本仓库](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills) |
| ⭐ [00.3](skills/00.3-Full-empirical-analysis-skill_R/) | **Full Empirical · R** 📗 | tidyverse · `fixest` · `did` · `HonestDiD` + Quarto 渲染 | [→](docs/CONTENT_ZH.md#skill-00-3) | [⭐ 本仓库](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills) |
| [01](skills/01-lishix520-academic-paper-skills/) | academic-paper-skills | 大纲 → 手稿写作 + 7 维审稿人模拟 | [→](docs/CONTENT_ZH.md#skill-01) | [lishix520/academic-paper-skills](https://github.com/lishix520/academic-paper-skills) |
| [02](skills/02-luwill-research-skills/) | research-skills | 医学影像综述、提案、论文转幻灯片 | [→](docs/CONTENT_ZH.md#skill-02) | [luwill/research-skills](https://github.com/luwill/research-skills) |
| [03](skills/03-K-Dense-AI-claude-scientific-skills/) | scientific-skills | 假设生成 + 28 个科学数据库 | [→](docs/CONTENT_ZH.md#skill-03) | [K-Dense-AI/claude-scientific-skills](https://github.com/K-Dense-AI/claude-scientific-skills) |
| [04](skills/04-K-Dense-AI-claude-scientific-writer/) | scientific-writer | 引用管理 + 科学写作 | [→](docs/CONTENT_ZH.md#skill-04) | [K-Dense-AI/claude-scientific-writer](https://github.com/K-Dense-AI/claude-scientific-writer) |
| [05](skills/05-kthorn-research-superpower/) | research-superpower | 系统化检索、筛选与引文溯源 | [→](docs/CONTENT_ZH.md#skill-05) | [kthorn/research-superpower](https://github.com/kthorn/research-superpower) |
| [06](skills/06-fuhaoda-stats-paper-writing/) | stats-paper-writing | 端到端 LaTeX 统计论文写作 | [→](docs/CONTENT_ZH.md#skill-06) | [fuhaoda/stats-paper-writing-agent-skills](https://github.com/fuhaoda/stats-paper-writing-agent-skills) |
| [07](skills/07-Orchestra-Research-AI-Research-SKILLs/) | AI-Research-SKILLs | 发表级 ML 图表、LaTeX、引文核验 | [→](docs/CONTENT_ZH.md#skill-07) | [Orchestra-Research/AI-Research-SKILLs](https://github.com/Orchestra-Research/AI-Research-SKILLs) |
| [08](skills/08-ndpvt-web-latex-document-skill/) | latex-document-skill | 创建 / 编译任意 LaTeX 文档为 PDF | [→](docs/CONTENT_ZH.md#skill-08) | [ndpvt-web/latex-document-skill](https://github.com/ndpvt-web/latex-document-skill) |
| [09](skills/09-meleantonio-awesome-econ-ai-stuff/) | awesome-econ-ai | Python 面板数据分析（`linearmodels`） | [→](docs/CONTENT_ZH.md#skill-09) | [meleantonio/awesome-econ-ai-stuff](https://github.com/meleantonio/awesome-econ-ai-stuff) |
| [10](skills/10-Jill0099-causal-inference-mixtape/) | causal-inference-mixtape | DID / IV / RDD / SCM 模板（Cunningham） | [→](docs/CONTENT_ZH.md#skill-10) | [Jill0099/causal-inference-mixtape](https://github.com/Jill0099/causal-inference-mixtape) |
| [11](skills/11-James-Traina-compound-science/) | compound-science | 面向定量社会科学的贝叶斯估计 | [→](docs/CONTENT_ZH.md#skill-11) | [James-Traina/compound-science](https://github.com/James-Traina/compound-science) |
| [12](skills/12-pedrohcgs-claude-code-my-workflow/) | claude-code-my-workflow | 提交 → PR → 合并的研究工作流（Emory） | [→](docs/CONTENT_ZH.md#skill-12) | [pedrohcgs/claude-code-my-workflow](https://github.com/pedrohcgs/claude-code-my-workflow) |
| [13](skills/13-scunning1975-MixtapeTools/) | MixtapeTools | Cunningham 的因果推断工具集与讲义 | [→](docs/CONTENT_ZH.md#skill-13) | [scunning1975/MixtapeTools](https://github.com/scunning1975/MixtapeTools) |
| [14](skills/14-luischanci-claude-code-research-starter/) | research-starter | R 中的 IV / DiD / RDD，含完整诊断 | [→](docs/CONTENT_ZH.md#skill-14) | [luischanci/claude-code-research-starter](https://github.com/luischanci/claude-code-research-starter) |
| [15](skills/15-Felpix-Studios-social-science-research/) | social-science-research | R 或 Python 端到端数据分析 | [→](docs/CONTENT_ZH.md#skill-15) | [Felpix-Studios/social-science-research](https://github.com/Felpix-Studios/social-science-research) |
| [16](skills/16-hsantanna88-clo-author/) | clo-author | 多代理数据分析（R / Stata / Python） | [→](docs/CONTENT_ZH.md#skill-16) | [hsantanna88/clo-author](https://github.com/hsantanna88/clo-author) |
| [17](skills/17-DAAF-Contribution-Community-daaf/) | DAAF | 安全意识代理框架（32 条 deny rule） | [→](docs/CONTENT_ZH.md#skill-17) | [DAAF-Contribution-Community/daaf](https://github.com/DAAF-Contribution-Community/daaf) |
| [18](skills/18-jusi-aalto-stata-accounting-research/) | stata-accounting | 来自 126 篇 *JAR* 论文的实测 Stata 范式 | [→](docs/CONTENT_ZH.md#skill-18) | [jusi-aalto/stata-accounting-research](https://github.com/jusi-aalto/stata-accounting-research) |
| [19](skills/19-CuellarC05-vera-economic-intelligence/) | vera-economic-intelligence | 经济情报 / 政策研究情报工作流 | [→](docs/CONTENT_ZH.md#skill-19) | [CuellarC05/vera-economic-intelligence](https://github.com/CuellarC05/vera-economic-intelligence) |
| [20](skills/20-wenddymacro-python-econ-skill/) | python-econ-skill | DSGE / HANK 与定量经济计算 | [→](docs/CONTENT_ZH.md#skill-20) | [wenddymacro/python-econ-skill](https://github.com/wenddymacro/python-econ-skill) |
| [21](skills/21-claesbackman-AI-research-feedback/) | AI-research-feedback | 用 AI 同行评审生成结构化反馈 | [→](docs/CONTENT_ZH.md#skill-21) | [claesbackman/AI-research-feedback](https://github.com/claesbackman/AI-research-feedback) |
| [22](skills/22-christopherkenny-skills/) | christopherkenny-skills | 面向 Quarto（`.qmd`）的 APSA 风格检查器 | [→](docs/CONTENT_ZH.md#skill-22) | [christopherkenny/skills](https://github.com/christopherkenny/skills) |
| [23](skills/23-Learning-Bayesian-Statistics-baygent-skills/) | baygent | 带护栏的 PyMC / Arviz 贝叶斯工作流 | [→](docs/CONTENT_ZH.md#skill-23) | [Learning-Bayesian-Statistics/baygent-skills](https://github.com/Learning-Bayesian-Statistics/baygent-skills) |
| [24](skills/24-Imbad0202-academic-research-skills/) | academic-research-skills | 5 审稿人多视角论文评审 | [→](docs/CONTENT_ZH.md#skill-24) | [Imbad0202/academic-research-skills](https://github.com/Imbad0202/academic-research-skills) |
| [25](skills/25-HosungYou-Diverga/) | Diverga | 研究问题精炼器（抗模式坍缩） | [→](docs/CONTENT_ZH.md#skill-25) | [HosungYou/Diverga](https://github.com/HosungYou/Diverga) |
| [26](skills/26-Data-Wise-scholar/) | scholar | 统计算法设计与文档 | [→](docs/CONTENT_ZH.md#skill-26) | [Data-Wise/claude-plugins](https://github.com/Data-Wise/claude-plugins) |
| [27](skills/27-dariia-m-my_claude_skills/) | my_claude_skills | 经济学摘要写作指南 | [→](docs/CONTENT_ZH.md#skill-27) | [dariia-m/my_claude_skills](https://github.com/dariia-m/my_claude_skills) |
| [28](skills/28-maxwell2732-paper-replicate-agent-demo/) | paper-replicate-agent | 论文复现代理演示 | [→](docs/CONTENT_ZH.md#skill-28) | [maxwell2732/paper-replicate-agent-demo](https://github.com/maxwell2732/paper-replicate-agent-demo) |
| [29](skills/29-quarcs-lab-project20XXy/) | project20XXy | 可复现手稿 + notebook 项目 | [→](docs/CONTENT_ZH.md#skill-29) | [quarcs-lab/project20XXy](https://github.com/quarcs-lab/project20XXy) |
| [30](skills/30-zirui-song-claude-skills/) | zirui-song-claude-skills | Zirui Song 的研究辅助 Claude 技能集 | [→](docs/CONTENT_ZH.md#skill-30) | [zirui-song/claude-skills](https://github.com/zirui-song/claude-skills) |
| [31](skills/31-thalysandratos-claude-code-skills/) | claude-code-skills | Python 面板数据分析 | [→](docs/CONTENT_ZH.md#skill-31) | [thalysandratos/claude-code-skills](https://github.com/thalysandratos/claude-code-skills) |
| [32](skills/32-dylantmoore-stata-skill/) | stata-skill | 高性能 Stata C/C++ 插件 | [→](docs/CONTENT_ZH.md#skill-32) | [dylantmoore/stata-skill](https://github.com/dylantmoore/stata-skill) |
| [33](skills/33-Galaxy-Dawn-claude-scholar/) | claude-scholar | 研究全生命周期：选题 → 综述 → 实验 → 审稿回复 | [→](docs/CONTENT_ZH.md#skill-33) | [Galaxy-Dawn/claude-scholar](https://github.com/Galaxy-Dawn/claude-scholar) |
| [34](skills/34-andrehuang-research-companion/) | research-companion | 头脑风暴、评估并决策研究方向 | [→](docs/CONTENT_ZH.md#skill-34) | [andrehuang/research-companion](https://github.com/andrehuang/research-companion) |
| [35](skills/35-bahayonghang-academic-writing-skills/) | academic-writing-skills | 面向投稿场所的工业 AI 文献研究 | [→](docs/CONTENT_ZH.md#skill-35) | [bahayonghang/academic-writing-skills](https://github.com/bahayonghang/academic-writing-skills) |
| [36](skills/36-taoyunudt-literature-review-skill/) | literature-review-skill | 完整文献综述工作流（中文） | [→](docs/CONTENT_ZH.md#skill-36) | [taoyunudt/literature-review-skill](https://github.com/taoyunudt/literature-review-skill) |
| [37](skills/37-IlanStrauss-ai-skills/) | IlanStrauss-ai-skills | Ilan Strauss 经济学研究 AI 工作流 | [→](docs/CONTENT_ZH.md#skill-37) | [IlanStrauss/ai-skills](https://github.com/IlanStrauss/ai-skills) |
| [38](skills/38-peternka-academic-proofreader/) | academic-proofreader | 学术校对 | [→](docs/CONTENT_ZH.md#skill-38) | [peternka/academic_proofreader](https://github.com/peternka/academic_proofreader) |
| [39](skills/39-vincentarelbundock-marginaleffects/) | marginaleffects | 预测、斜率与比较（R / Python） | [→](docs/CONTENT_ZH.md#skill-39) | [vincentarelbundock/marginaleffects](https://github.com/vincentarelbundock/marginaleffects) |
| [40](skills/40-py-econometrics-pyfixest/) | pyfixest | Python 中的快速固定效应估计 | [→](docs/CONTENT_ZH.md#skill-40) | [py-econometrics/pyfixest](https://github.com/py-econometrics/pyfixest) |
| [41](skills/41-sticerd-eee-sewage-econometrics-check/) | sewage-econometrics-check | 10 项复现包审计 | [→](docs/CONTENT_ZH.md#skill-41) | [sticerd-eee/sewage](https://github.com/sticerd-eee/sewage) |
| [42](skills/42-wanshuiyin-ARIS/) | ARIS | 自主「research-in-sleep」代理，端到端 | [→](docs/CONTENT_ZH.md#skill-42) | [wanshuiyin/Auto-claude-code-research-in-sleep](https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep) |
| [43](skills/43-wentorai-research-plugins/) | research-plugins | 478 个研究插件：数据可视化、领域、基础设施 | [→](docs/CONTENT_ZH.md#skill-43) | [wentorai/research-plugins](https://github.com/wentorai/research-plugins) |
| [44](skills/44-matsuikentaro1-humanizer_academic/) | humanizer_academic | 为医学/学术手稿去 AI 味（34 类模式） | [→](docs/CONTENT_ZH.md#skill-44) | [matsuikentaro1/humanizer_academic](https://github.com/matsuikentaro1/humanizer_academic) |
| [45](skills/45-stephenturner-skill-deslop/) | deslop | 去除 AI 写作痕迹（5 维评分） | [→](docs/CONTENT_ZH.md#skill-45) | [stephenturner/skill-deslop](https://github.com/stephenturner/skill-deslop) |
| [46](skills/46-hardikpandya-stop-slop/) | stop-slop | 三层 AI 痕迹检测与改写 | [→](docs/CONTENT_ZH.md#skill-46) | [hardikpandya/stop-slop](https://github.com/hardikpandya/stop-slop) |
| [47](skills/47-conorbronsdon-avoid-ai-writing/) | avoid-ai-writing | 审计 → 改写 → 二次审计 AI 味（留痕） | [→](docs/CONTENT_ZH.md#skill-47) | [conorbronsdon/avoid-ai-writing](https://github.com/conorbronsdon/avoid-ai-writing) |
| ⭐ [48](skills/48-de-AIGC-skills/) | **de-AIGC-skills** 🇨🇳🇬🇧 | 中英双语学术降 AIGC + 去水印层（Turnitin AI / GPTZero / 知网 / 万方 · 隐藏字符 / C2PA / docx 元数据） | [→](docs/CONTENT_ZH.md#skill-48) | [⭐ 本仓库](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills) |
| [49](skills/49-voidborne-d-humanize-chinese/) | humanize-chinese | 检测并人性化 AI 生成的中文文本 | [→](docs/CONTENT_ZH.md#skill-49) | [swaylq/humanize-chinese](https://github.com/swaylq/humanize-chinese) |
| ⭐ [50](skills/50-brycewang-aer-skills/) | **AER-skills** 📕 | Top-5 经济学投稿套件：识别 → 稳健性 → R&R | [→](docs/CONTENT_ZH.md#skill-50) | [brycewang-stanford/AER-skills](https://github.com/brycewang-stanford/AER-skills) |
| [51](skills/51-pymc-labs-CausalPy/) | CausalPy | 贝叶斯准实验（PyMC Labs） | [→](docs/CONTENT_ZH.md#skill-51) | [pymc-labs/CausalPy](https://github.com/pymc-labs/CausalPy) |
| [52](skills/52-keemanxp-slr-prisma/) | slr-prisma | 系统文献综述，PRISMA 2020 | [→](docs/CONTENT_ZH.md#skill-52) | [keemanxp/slr-prisma](https://github.com/keemanxp/slr-prisma) |
| [53](skills/53-keemanxp-thematic-analysis-skill/) | thematic-analysis | Braun & Clarke 六阶段定性主题分析 | [→](docs/CONTENT_ZH.md#skill-53) | [keemanxp/thematic-analysis-skill](https://github.com/keemanxp/thematic-analysis-skill) |
| [54](skills/54-scdenney-open-science-skills/) | open-science-skills | 引用一致性、DOI 与论据支撑审计 | [→](docs/CONTENT_ZH.md#skill-54) | [scdenney/open-science-skills](https://github.com/scdenney/open-science-skills) |
| [55](skills/55-ab604-claude-code-r-skills/) | r-skills | R 中用 `brms` 做贝叶斯推断 | [→](docs/CONTENT_ZH.md#skill-55) | [ab604/claude-code-r-skills](https://github.com/ab604/claude-code-r-skills) |
| [56](skills/56-hanlulong-econ-writing-skill/) | econ-writing-skill | 综合 50+ 顶级指南的经济学写作 | [→](docs/CONTENT_ZH.md#skill-56) | [hanlulong/econ-writing-skill](https://github.com/hanlulong/econ-writing-skill) |
| [57](skills/57-dgunning-edgartools/) | edgartools | 查询与分析 SEC 文件 | [→](docs/CONTENT_ZH.md#skill-57) | [dgunning/edgartools](https://github.com/dgunning/edgartools) |
| [58](skills/58-charlescoverdale-econstack/) | econstack | 政策简报（UK GES / AU Treasury） | [→](docs/CONTENT_ZH.md#skill-58) | [charlescoverdale/econstack](https://github.com/charlescoverdale/econstack) |
| [59](skills/59-shiquda-openalex-skill/) | openalex-skill | 通过 OpenAlex 查询 2.4 亿+ 学术作品 | [→](docs/CONTENT_ZH.md#skill-59) | [shiquda/openalex-skill](https://github.com/shiquda/openalex-skill) |
| [60](skills/60-regisely-superpapers/) | superpapers | 综合性实证研究支持套件 | [→](docs/CONTENT_ZH.md#skill-60) | [regisely/superpapers](https://github.com/regisely/superpapers) |
| [61](skills/61-phdemotions-research-methods/) | research-methods | 与预注册匹配的验证性检验 | [→](docs/CONTENT_ZH.md#skill-61) | [phdemotions/research-methods](https://github.com/phdemotions/research-methods) |
| [62](skills/62-PHY041-claude-skill-citation-checker/) | citation-checker | 对照 CrossRef / S2 / OpenAlex 核验引用 | [→](docs/CONTENT_ZH.md#skill-62) | [PHY041/claude-skill-citation-checker](https://github.com/PHY041/claude-skill-citation-checker) |
| [63](skills/63-tondevrel-scientific-agent-skills/) | scientific-agent-skills | DoWhy 识别–估计–反驳框架 | [→](docs/CONTENT_ZH.md#skill-63) | [tondevrel/scientific-agent-skills](https://github.com/tondevrel/scientific-agent-skills) |
| [64](skills/64-tmonk-mcp-stata/) | mcp-stata | 20 个 Stata 因果推断与复现 skill | [→](docs/CONTENT_ZH.md#skill-64) | [tmonk/mcp-stata](https://github.com/tmonk/mcp-stata) |
| [65](skills/65-game-theory-paper-writer/) | game-theory-paper-writer | 生成并压力测试博弈论论文 | [→](docs/CONTENT_ZH.md#skill-65) | [本仓库 PR #17](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/pull/17) |
| [66](skills/66-zheng-siyao-empirical-research-skills/) | empirical-research-skills | 面向大型面板的 R 性能优化 | [→](docs/CONTENT_ZH.md#skill-66) | [SiyaoZheng/ai4ss-skills](https://github.com/SiyaoZheng/ai4ss-skills) |
| [67](skills/67-econfin-workflow-toolkit/) | econfin-workflow-toolkit | 中国公司金融实证工作流，从提案到论文 | [→](docs/CONTENT_ZH.md#skill-67) | [本仓库 PR #22](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/pull/22) |
| [68](skills/68-research-productivity-skills/) | research-productivity-skills | 论文检索、SSRN、DOI 查询、下载 | [→](docs/CONTENT_ZH.md#skill-68) | [本仓库 PR #21](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/pull/21) |
| ⭐ [69](skills/69-Paper-WorkFlow/) | **Paper-WorkFlow** 🧭 | 元编排器，串起整个社会科学论文流水线 | [→](docs/CONTENT_ZH.md#skill-69) | [brycewang-stanford/Paper-WorkFlow](https://github.com/brycewang-stanford/Paper-WorkFlow) |
| [70](skills/70-ssci-polish/) | ssci-polish ✍️ | SSCI / SCI 英文论文语言润色（语法、可读性、学术语气） | [→](docs/CONTENT_ZH.md#skill-70) | [⭐ 本仓库](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills) |
| ⭐ [71](skills/71-brycewang-lit-review-agent-tools/) | **lit-review-agent-tools** 🔍 | 文献综述工具选型 + 一键安装运行（MinerU / PaperQA2 / ASReview / STORM / MCP 服务器） | [→](docs/CONTENT_ZH.md#skill-71) | [brycewang-stanford/lit-review-agent-tools](https://github.com/brycewang-stanford/lit-review-agent-tools) |
| ⭐ [72](skills/72-kaggle-research/) | **Kaggle Research** 🧪 | 通过官方 CLI 安全检索 Kaggle 资源、限界下载公开数据并保留审计证据 | [→](docs/CONTENT_ZH.md#skill-72) | [⭐ 本仓库](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills) |
| ⭐ [73](skills/73-brycewang-p-hacking-skills/) | **p-hacking-skills** 🔬 | p-hacking 体检室：在真实效应为零的数据上仪表化规格搜索，给出诚实 p 值、p 曲线检测与预注册免疫（教学与审计用） | [→](docs/CONTENT_ZH.md#skill-73) | [brycewang-stanford/p-hacking-skills](https://github.com/brycewang-stanford/p-hacking-skills) |

> **想看更详细的描述（主题分类、字段、统计）？** 见 [`docs/CONTENT_ZH.md`](docs/CONTENT_ZH.md) 中标注 `#skill-NN` 锚点的同一张表 —— 它是每个合集的完整描述所在的扩展正文。

## 📈 项目历程

自 2026-04 首次发布以来的主干里程碑（完整提交记录见 [Commits](https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/commits/main) 与 [`CHANGELOG.md`](CHANGELOG.md)）：

```mermaid
---
config:
  gitGraph:
    rotateCommitLabel: false
---
gitGraph TB:
   commit id: "2026-04 首次发布"
   branch community
   commit id: "2026-05 首个社区 PR"
   checkout main
   merge community
   commit id: "2026-05 更名 AERS"
   commit id: "2026-06 插件市场"
   commit id: "2026-06 全库路由器"
   commit id: "2026-07 首个 tag" tag: "v2026.07"
   branch kaggle
   commit id: "2026-07 Kaggle 集成"
   checkout main
   merge kaggle
   commit id: "2026-08 de-AIGC 双语"
   commit id: "2026-08 来源链接全覆盖"
   branch evidence
   commit id: "2026-08 aers-score CLI"
   commit id: "2026-08 外部成绩单"
   checkout main
   merge evidence
   commit id: "2026-08 结构估计 = 方法族 18"
   commit id: "2026-08 NSW 基准从引用变推导"
   commit id: "2026-09 数据 → Word 全稿路线"
   commit id: "2026-09 de-AIGC 去水印层"
   commit id: "2026-09 分层检索 + 路由评测"
   commit id: "2026-09 StatsPAI 置顶"
   branch phacking
   commit id: "2026-09 p-hacking 体检室（合集 73）"
   checkout main
   merge phacking
```

<div align="center">

<a href="https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/stargazers">
 <picture>
   <source media="(prefers-color-scheme: dark)" srcset="images/star-history-dark.svg" />
   <source media="(prefers-color-scheme: light)" srcset="images/star-history.svg" />
   <img alt="Star History Chart" src="images/star-history.svg" width="600" />
 </picture>
</a>

<sub>Star 增长曲线（非提交数）· 由 <a href="scripts/build-star-history.py"><code>scripts/build-star-history.py</code></a> 从 GitHub API 生成并提交入库</sub>

</div>

如果 AERS 对你的工作有帮助，请**引用它**（[CITATION.cff](CITATION.cff)）并**点个 Star**，让更多研究者看到。

---

<div align="center">

**AI 是放大器，不是替代品。它替你做最耗时的"搬砖"，你保留最核心的"判断"。**

<br/>

<table>
  <tr>
    <td align="center">
      <a href="https://github.com/brycewang-stanford/StatsPAI"><img src="images/statspai-logo.png" alt="StatsPAI" width="260" /></a>
    </td>
    <td width="40"></td>
    <td align="center">
      <img src="images/stanford-reap-logo.png" alt="Stanford REAP" width="320" />
    </td>
  </tr>
</table>

<sub><strong>Stanford REAP × StatsPAI</strong> · 实证研究 AI 工具的学术工业级产品</sub>

<br/>

<table>
 <tr>
    <td align="center">
      <a href="https://copaper.ai"><img src="images/copaper-qrcode.png" alt="扫码访问 copaper.ai" width="180" /></a><br/>
      <strong>扫码访问 <a href="https://copaper.ai">copaper.ai</a></strong>
    </td>
    <td align="center">
      <img src="images/copaper-wechat.jpg" alt="CoPaper.AI 公众号" width="180" /><br/>
      <strong>关注公众号「CoPaper.AI」</strong>
    </td>
  </tr>
</table>

内置 20 个方法论 skill · 20 分钟完成实证论文 · 自研 <a href="https://github.com/brycewang-stanford/StatsPAI"><strong>StatsPAI</strong></a>（900+ 函数 / MIT 开源）

</div>

---
*Source: https://skills.yangsir.net/skill/gh-auto-empirical-research-skills*
*Markdown mirror: https://skills.yangsir.net/api/skill/gh-auto-empirical-research-skills/markdown*