Top 10 Skills & Tools for Social Science Research!

1. Auto-Empirical-Research-Skills — Stanford team’s self-developed 23k+ empirical research Agent Skills family bucket
   https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills
2. StatsPAI — Agent-native causal econometrics Python library with unified API + structured output
   https://github.com/brycewang-stanford/StatsPAI
3. Awesome-Journal-Skills — Journal-specific submission skill package (AER / QJE / Nature, etc.)
   https://github.com/brycewang-stanford/Awesome-Journal-Skills
4. AER-Skills — Top-tier journal submission skill stack for AER and AEJ series
   https://github.com/brycewang-stanford/AER-Skills
5. Auto-Research-Skills — Autonomous research skills and Agent curated library
   https://github.com/brycewang-stanford/Auto-Research-Skills
6. statsmodels — Classic statistical modeling and econometrics library for Python
   https://github.com/statsmodels/statsmodels
7. dowhy — Core causal inference library (DoWhy), graphical models + hypothesis testing
   https://github.com/py-why/dowhy
8. causalml — Uber’s uplift modeling and causal ML toolkit
   https://github.com/uber/causalml
9. EconML — Heterogeneous causal effect (CATE) machine learning toolkit
   https://github.com/py-why/EconML
10. python-causality-handbook — Causal Inference for the Brave and True practical handbook
    https://github.com/matheusfacure/python-causality-handbook

⭐ Star the first 5 self-developed repos from brycewang-stanford, especially if you want to write papers and do empirical research with AI Agents — your productivity will skyrocket.

brycewang-stanford/Auto-Empirical-Research-Skills

Source: https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills

Auto-Empirical Research Skills (AERS)

🌐 Languages: English | 简体中文 (default) | 繁體中文 | 日本語 | 한국어

Stanford REAP × CoPaper.AI · An academic-grade industrial product of empirical research AI tools

Built by the Stanford Empirical Research Methodology team, covering the complete workflow from data cleaning to top-journal submission

Awesome (https://awesome.re)
GitHub stars (https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills)
License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/)

PRs Welcome
Validate catalog (https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills/actions/workflows/validate-catalog.yml)
OpenSSF Scorecard (https://scorecard.dev/viewer/?uri=github.com/brycewang-stanford/Auto-Empirical-Research-Skills)
Security audit: 52/52 CLEAN

Powered by StatsPAI (https://github.com/brycewang-stanford/StatsPAI)

First: The entire repo is an AERS Router Skill

This repository supports importing the entire root directory as a single skill into Codex, CodeBuddy, Claude Code, or similar IDEs. The root directory’s SKILL.md registers as auto-empirical-research-skills, which routes to the appropriate vendored sub-skill based on your research task — instead of loading all 69 collections and 1,145 skills into context at once.

• Recommended: Import the whole repo: Select the repository root, use agents/openai.yaml and the root SKILL.md to register a catalog router; the agent first queries catalog/skills.json or docs/SKILL_CATALOG.md, then reads only the target sub-skill.
• Recommended for single skill import: If your IDE requires “one directory per skill,” copy the subdirectory that directly contains a SKILL.md, e.g., skills/50-brycewang-aer-skills/skills/aer-workflow/, and do not mistakenly register the entire repo as 1,145 independent skills recursively.
• Installation details: See docs/INSTALL.md. The Chinese README is the default entry; the English entry is preserved in README-en.md.

Complete Overview of All 69 Skill Collections

Open the repo → see the entire library.
All 69 collections · 1,145 skills, numbered from 00 to 69, each vendored into this repo (not just external links), tracked by catalog/skills.json. Click any row to open its folder.

⭐ = Self-developed skill by the Stanford REAP × CoPaper.AI team; others are curated, security-audited community works.

Legend: 🚀 Full workflow & orchestrator · 🎯 Causal inference & econometrics · 📚 Literature & research design · ✍️ Writing, editing & de-AIGC · 📑 Citation, reproduction & peer review · 🛠️ Data, tools & infrastructure

⭐ Our handcrafted backbone: StatsPAI (causal engine) · explicit Python / Stata / R full workflow ports · AER-skills (Top-5 submission kit) · chinese-de-aigc · Paper-WorkFlow (meta-orchestrator). They are the backbone of AERS — see full comparison in Flagship Pipeline Skills ↓. Prefer to browse by use case? See Same 69 collections, grouped by purpose ↓.

An empirical research-oriented professional-grade Agent Skills distribution.
Not a marketing list — this repo owns and catalogs 1,145 skills, plus a set of numerical benchmarks + evaluation suite + security audit + CI that locks down quality, overlaid with a curated map covering 23,000+ skills / 119 repos in the ecosystem.
AERS is two things at once: (1) a small set of self-developed flagship skills that can run the complete empirical pipeline — data cleaning → identification → estimation → robustness → tables/figures → submission-ready draft; (2) a curated, security-controlled catalog that organizes the entire empirical research skill ecosystem by research process stage. Our differentiation is not in quantity, but in the fact that the flagship skills’ behavior is verified against known answers, not just claimed verbally.

Renamed. This project was formerly Awesome Agent Skills for Empirical Research. GitHub will automatically redirect old addresses, but please update your local remote:

git remote set-url origin https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git  

Table of Contents

• Complete Overview of All 69 Skill Collections (full 00 → 69 index)
• 69 Collections · Grouped by Purpose
• What You Actually Get (Exact Numbers)
• Verify in 2 Minutes
• Why You Can Trust It — Three Layers of Credibility Anchors
• Flagship Pipeline Skills
• Start Here — Pick a Skill in 30 Seconds
• Why It’s More Than Just a Stack of 23K Skills
• Browse the Full Panorama
• Tool Catalog (tools/)
• By Research Workflow
• Comprehensive Skill Suites
• Reduce AIGC Detection Rate & Academic De-AI-fy
• Multi-Agent Systems · MCP Servers · Platforms · Learning Resources
• Security Scan
• Changelog
• Contributing & Citation

69 Collections · Grouped by Purpose

The same 69 collections · 1,145 skills as the sequential index above ↑ — here reordered by research purpose so you can jump directly to your current stage. ⭐ = Self-developed (Stanford REAP × CoPaper.AI); others are curated, security-audited community works.

🚀 Full Workflow Flagships & Orchestrators — One call runs the entire empirical loop

🎯 Causal Inference & Econometrics — The methodological core of AERS

📚 Literature, Reading & Research Design — From question to evidence base

✍️ Writing, Editing & De-AIGC — From draft to submission-ready

📑 Reproduction, Citation & Peer Review — Quality assurance and transparency

🛠️ Data, Tools & Infrastructure — The foundation layer