Make my research analysis reproducible and shareable
Organizes my project's data, code, and documentation so others (and future me) can rerun the analysis and get the same results.
The prompt
Help me make my research analysis reproducible before I publish or share it.
Language and tools: [e.g. "R with RStudio", "Python and Jupyter", "Stata", "MATLAB"]
Current project structure: [PASTE a folder listing or describe it]
Data: [raw sources, size, whether it can be shared publicly, any personal data]
How the analysis runs today: [e.g. "I run scripts in order by hand", "one big notebook"]
Where I plan to share: [e.g. "GitHub plus Zenodo", "OSF", "journal supplement"]
Give me:
1. A recommended folder structure (raw data kept read-only, processed data, code, outputs, docs) and how to move my current files into it.
2. A way to run everything in order from raw data to final figures and tables with one command (e.g. a main script, Makefile, or targets/snakemake), with a starter version.
3. Environment capture: how to record package versions (e.g. renv, requirements.txt, conda environment file) and set random seeds.
4. A README template: purpose, how to run, software requirements, data access instructions, expected outputs mapped to paper figures and tables, license, and how to cite.
5. A pre-release checklist: run on a clean machine, remove hard-coded paths and credentials, check that no personal or restricted data is included, and assign a DOI.
Keep advice proportionate to a single research project, not a software product.
Highlighted [brackets] are blanks to replace with your own details.
Fill in the blanks and use this promptMore Software Development prompts
What to fill in
- PASTE a folder listing or describe it
- e.g. "I run scripts in order by hand", "one big notebook"
- e.g. "GitHub plus Zenodo", "OSF", "journal supplement"
How this prompt is built
This prompt follows the Few-Shot structure (Few-shot prompting). A few-shot prompt includes one or more worked examples of the input and the output you want, so the model can copy the pattern instead of guessing it.
Related prompts
- Analyze logs to find what went wrongReads logs to identify errors, patterns, and the likely root cause.
- Get a beginner-friendly code review of my first projectA ranked set of the five improvements that matter most, explained simply, plus what you did well and what to learn next.
- Audit a codebase for bugs, risks, and tech debtA systematic review of a repository with prioritized findings and a remediation plan.
- Create a test plan and test casesA risk-based test strategy with test cases for functional, integration, and end-to-end testing.