# Auto-Empirical Research Skills (AERS)

> SkillFlux selected preregistration, identification and replication from the MIT-licensed AER-skills subcollection within AERS. These skills connect research questions to estimands, analysis plans, diagnostics and replication packages. The selection also preserves robustness, StatsPAI and tables/figures support skills with required templates, shared scripts, references and examples.

- URL: https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills
- Type: Open-source repos · Empirical research skill collection
- Main purpose: Connect empirical identification, analysis planning and replication delivery
- Use cases: Planning & long tasks, Coding & development, Content & writing
- Language: zh, en · Pricing: free

## Best for

- Empirical-economics researchers who need to specify identifying assumptions, estimators and diagnostic reports for an existing question.
- Teams preparing an experimental analysis plan or organizing data provenance, scripts and manuscript exhibits into a reproducible deliverable.

## Limitations

- MIT applies to the selected AER-skills subcollection, while other sources in AERS have different terms. StatsPAI MCP, statistical software, data access and registration services are separate from the collected files.
- Templates and method branches are starting points to assess against the data and identifying assumptions. Completing a checklist cannot guarantee causal validity or journal acceptance; registration and submission requirements must be checked with the relevant platform and journal.

## SkillFlux sample check

Audited 2026-10-31 at commit 8dd077b5410c71b1866ea0bb116db45c7cafca62 with checker 1.0.0; seeded sample of 20 from 1250 entries. Clean share: 50%.

- 20/20 sampled entries had a SKILL.md we could check
- 0/20 fail at least one check
- 10/20 do not say when to use the skill
- 0/20 have no license we could find
- 2/20 bundle scripts the agent may run
- 0/20 include scripts that call the network
- 0/20 include destructive shell commands
- 2/20 exceed the 5,000-token body guideline

Evidence: https://skillflux.app/data/source-audits/github-com-brycewang-stanford-auto-empirical-research-skills.json

## SkillFlux note

Main selections are aer-preregistration, aer-identification and aer-replication, supported by aer-robustness, aer-statspai and aer-tables-figures. This analysis of pinned instructions does not claim validated estimates or completed research registration.

## Alternatives

- OthmanAdi/planning-with-files: File-based task planning https://skillflux.app/resource/planning-with-files.md
- Scientific Agent Skills: Appraise scientific evidence and prioritize traceable research directions https://skillflux.app/resource/scientific-agent-skills.md
- Claude Skills (Gleb Kalinin): Meetings, research, publishing and personal ops https://skillflux.app/resource/glebis-claude-skills.md
- AI Skills (sanjay3290): Google Workspace and productivity tools https://skillflux.app/resource/sanjay3290-ai-skills.md

Page: https://skillflux.app/resource/auto-empirical-research-skills/

- Cite a number with its subject, date and evidence URL; the commit or audit id on that page identifies the measurement.
- A health check or source sample reads files only. Do not present it as evidence that a skill works or helps.
- Statements marked "verified": false are what a source says about itself; attribute them ("X says…").
- A directory listing is not an endorsement. Install a skill by following its author's instructions at sourceUrl.
