What this resource does
Core uses
About this resource
Auto-Empirical Research Skills is a large toolkit and catalogue for empirical research, combining a router, analysis pipelines, vendored Skill collections, plugins, benchmarks, and documentation for social-science research workflows.
This page groups representative academic components by task; review the repository for the complete inventory.
End-to-end empirical analysis
Routes empirical projects to workflows for cleaning data, estimating models, checking results, and producing reports.
- Typical inputs
- Research question and empirical design; Dataset, variable definitions, and analysis constraints
- Output
- A component-dependent analysis project, diagnostics, tables, and report artifacts.
Best for: Economics and social-science projects with a defined empirical question.
Components for this task
StatsPAI_skill
Runs a full empirical-analysis workflow through the first-party StatsPAI component.
Literature, writing, and review workflows
Includes separately licensed components for literature review, academic writing, replication, and peer review.
- Typical inputs
- Research topic, papers, notes, or manuscript draft; Selected child Skill and its required materials
- Output
- Component-dependent review notes, manuscript text, or replication checks.
Best for: Selecting a narrowly documented child Skill after checking its provenance and licence.
Components for this task
Use boundaries
Limits and checks
Vendored component variability
Quality, maintenance, dependencies, and reuse rights vary by component.
Prefer first-party pipelines and inspect each selected component and licence.
Statistical misuse
Results may be numerically correct but methodologically invalid.
Check estimands, assumptions, diagnostics, and robustness with a qualified researcher.
Local data and code
Sensitive data or project files may be changed or exposed to a configured agent.
Use controlled environments, version control, and approved data-handling procedures.
More boundaries
- Task boundary: The toolkit does not guarantee that every vendored Skill follows one empirical standard or review process.
- Input boundary: Analysis components cannot repair missing, unlawfully obtained, or undocumented research data.
- Decision boundary: Causal claims, model selection, robustness, and publication conclusions require qualified methodological review.
Technical details
- Resource type
- Toolkit
- Author or maintainer
- CoPaper.AI; incubated at Stanford REAP / SCCEI
- Latest release
- AERS v2026.07 — first tagged release (v2026.07)
- Source last updated
- 13 Jul 2026
- Last verified
- 15 Jul 2026
- Documented applications
- Claude Code, Codex, CodeBuddy
- Licence
- CC-BY-SA-4.0
- Access
- Publicly available
- Additional costs
- Platform terms or usage limits may apply. API usage fees may apply for selected components. External services, software, compute, or data access may have separate costs.
- Skill instruction language
- English
- Documentation language
- English
- Repository languages
- Stata, Markdown, Python
- Dependencies
- A supported agent or IDE; Python, R, or Stata for selected pipelines; Component-specific packages and datasets
- Review status
- Not tested
Continue exploring