</>MCP Agents Market
Skill

Auto-Empirical-Research-Skills

by brycewang-stanford3.6kStataUpdated 2026-08-31

🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.

Claude CodeCodexCodeBuddy

Auto-Empirical-Research-Skills (AERS) is a comprehensive library of over 23,000 agent skills spanning 76 curated collections for empirical research in social sciences. Maintained by Stanford REAP and CoPaper.AI, it provides end-to-end pipeline automation covering research design, causal inference (DID/RD/IV/SCM), statistical estimation in Python/Stata/R, robustness testing, publication-quality tables/figures, academic writing, and AI-content detection for Chinese and English manuscripts. The toolkit includes flagship skills like StatsPAI (900+ functions for causal analysis), full empirical analysis stacks for three languages, de-AIGC tools for Turnitin/CNKI compliance, and a meta-orchestrator (Paper-WorkFlow) that chains the entire research lifecycle from idea to journal submission.

Key Features

76 skill collections totaling 1,096+ individual skills covering 9 research stages from topic refinement to submission
StatsPAI causal engine with 900+ functions for DID, RD, IV, synthetic control, double ML, and matching methods
Language-specific empirical analysis stacks for Python (pandas/pyfixest), Stata (reghdfe/csdid), and R (fixest/HonestDiD)
Bilingual de-AIGC skills for removing AI writing patterns detectable by Turnitin AI, GPTZero, CNKI, and Wanfang
AER-skills for top-5 economics journal submission with identification strategies and R&R workflows
Paper-WorkFlow meta-orchestrator that chains all 9 stages into one-command end-to-end automation
19 numeric benchmarks and 42 behavioral evaluation scenarios with pass/fail fixtures for rigor verification
Reproducibility audit tools (sewage-econometrics-check) performing 10-item replication package checks

Use Cases

  • 01Running complete empirical analysis pipelines from data cleaning to publication-ready tables in Python, Stata, or R
  • 02Conducting causal inference studies with difference-in-differences, regression discontinuity, or instrumental variables
  • 03Performing systematic literature reviews with PRISMA 2020 compliance and citation verification
  • 04Generating journal-quality econometric tables and figures with LaTeX embedding for top-tier submissions
  • 05Removing AI-generated content patterns from Chinese or English academic manuscripts for plagiarism checker compliance
  • 06Automating robustness checks including HonestDiD, sensitivity analysis, and pre-registration alignment tests

Related Skills

View more

Auto-Empirical-Research-Skills — FAQ

What is Auto-Empirical-Research-Skills?+

AERS is a curated collection of 23,000+ agent skills across 76 repositories designed for empirical research in social sciences, covering the complete workflow from research design to journal submission. It includes flagship tools from Stanford REAP and CoPaper.AI plus vetted community contributions.

How do I install AERS skills?+

For Claude Code v2.1+, use 'claude plugin marketplace add brycewang-stanford/Auto-Empirical-Research-Skills' then install specific skills. Alternatively, clone the repo and copy skill folders (containing SKILL.md) to .claude/skills/ in your project root or home directory for global access.

Which AI clients does this work with?+

AERS is primarily designed for Claude Code (v2.1+) and Codex, with skills structured as SKILL.md files that these agents auto-discover. Some collections like Paper-WorkFlow also support CodeBuddy and other MCP-compatible clients.

Do I need API keys or special prerequisites?+

Most skills require Python 3.8+ and standard scientific libraries (pandas, statsmodels, etc.). Some collections need Stata or R installed locally. Literature review skills (claude-scholar, openalex-skill) may require API keys for OpenAlex, Semantic Scholar, or CrossRef for full functionality.

Is Auto-Empirical-Research-Skills free to use?+

Yes, the repository is CC BY-SA 4.0 licensed for the catalog structure, with individual skills carrying their upstream licenses (mostly MIT/Apache-2.0). All 76 collections are freely available; see docs/LICENSE_AUDIT.md for per-skill license details.

How does the end-to-end pipeline work?+

The Paper-WorkFlow skill (#69) acts as a meta-orchestrator, chaining the 9 research stages (topic refinement → literature review → data acquisition → identification strategy → estimation → robustness → tables/figures → writing → de-AIGC) automatically. You can intervene at any stage, manually adjust methods or variables, and let the pipeline continue from that point.

How do I install Auto-Empirical-Research-Skills?+

Open the source repository on GitHub and follow its README. Auto-Empirical-Research-Skills is a skill — MCP Agents Market links you directly to the official repo.

Is Auto-Empirical-Research-Skills free?+

Auto-Empirical-Research-Skills is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches