open-science-skills
Design, run, and report transparent social-science research
All Skills in This Repository (11)
Pure Emerald Level Indicatorspre-registration-writing
Draft pre-analysis plans for experimental social science registries.
methods-reporting
Audit experimental social science methods sections against a 45-item checklist.
narrative-building
Draft scientific introductions and literature reviews from a research interest.
conjoint-design
Design and analyze conjoint experiments with attribute architecture and power analysis.
conjoint-diagnostics
Diagnose conjoint studies using a systematic checklist covering design, estimation, and validity.
text-classification
Classify social science text with LLMs using a defined codebook.
hypothesis-building
Formalize causal hypotheses into estimands for preregistration and analysis planning.
survey-design
Design survey instruments to minimize measurement error and bias in experimental social science research.
list-experiment
Plan, design, and diagnose list-experiment studies for sensitive prevalence surveys.
cross-national-design
Guide cross-national comparative survey experiment design with per-country power analyses.
topic-modeling
Select and document STM, LDA, or BERTopic topic-modeling approaches with preprocessing and diagnostics.
Frequently Asked Questions
FAQPage SchemaHow to install Open Science Skills?▼
Run `npx skills add scdenney/open-science-skills --all -g -y` in your terminal to install all skills globally.
What research tasks does Open Science Skills cover?▼
It covers the full research lifecycle: hypothesis building, survey and conjoint design, pre-registration, text classification, topic modeling, OCR of scanned documents, manuscript review, citation checking, and replication packaging.
Does it work with Claude Code and OpenAI Codex?▼
Yes. The library ships parallel versions for both platforms, invoked as /oss:skill-name in Claude Code and $skill-name in Codex.
Can it help with pre-registration and reporting standards?▼
Yes. Dedicated skills draft pre-analysis plans for OSF, AEA, or AsPredicted and audit methods sections against a 45-item checklist built from APSA, JARS-Quant, CONSORT, and DA-RT.
Do I need statistics expertise to use it?▼
No. Each skill explains the methodological reasoning in plain language, gives recommended defaults, and flags the decisions only you as the researcher can make.
Related Repositories in Education & Research
View All in Education & Research→learn-claude-code
Learn to build AI agent harnesses from scratch
ai-agents-for-beginners
Learn to build, deploy, and secure AI agents step by step
33-js-concepts
Master 33 essential JavaScript concepts with tested, verified guides