Skip to content
Oshki

How it works

From a rough idea to a submission-ready manuscript.

One project carries your study through every stage, with the right tools and reporting guideline at each one. Here’s the path, shown with one fictional study: sleep quality and exam stress in medical students.
Join the waitlist

Before stage 1 · optional

Start from an idea.

Intake turns a rough question into structured answers, the closest published studies, an honest verdict, and two or three designs to choose from. Already know your design? Skip straight to the template.

Intake — start from an idea

What do you want to find out?

Closest published studiesPubMed + OpenAlex

Illustrative preview — fictional study, not real data.

Stage 1 · Protocol

A protocol shaped by your guideline.

Objectives, questions, and design, structured around the reporting guideline your study type commits to — so you see what the guideline asks for while you work, not the night before submission.
  • Plain-language guidance for every stage
  • Tasks grouped by stage, completed by people

Sleep quality and exam stress in medical students

Cross-sectional surveySTROBEStage 3 of 5

Sleep quality and exam stress in medical students

Data collection84 / 120 responses

STROBE checklist

6 of 22 addressed

  • 1Title and abstract
  • 2Background / rationale
  • 3Objectives
  • 4Study design
  • 5Setting
  • 6Participants
  • 7Variables
  • 8Data sources / measurement

Illustrative preview — fictional study, not real data.

Stages 2–3 · Design & data collection

Build the instrument, then watch responses arrive.

A survey for cross-sectional studies, a case report form for cohort, case-control, and trial designs — or a dataset upload for secondary analysis.

Survey design — Sleep & exam stress

Q1Single choice

Year of study

Year 1Year 2Year 3Year 4+
Q2Likert · 5

Over the past month, how would you rate your sleep quality?

12345Very poorVery good
Mobile-first respondent page · anonymousShare linkQR code

Illustrative preview — fictional study, not real data.

Stage 4 · Analysis

The test your data calls for.

Tests suggested from your variable types and study design, with APA-style results tables you can copy into the manuscript.

Analysis — Sleep & exam stress

Data · 84 rows

idpsqiordstressord
R-001924
R-002517
R-0031229
R-004415
R-005821

Suggested test

Results

Illustrative preview — fictional study, not real data.

Stage 5 · Write-up

AI assists. You decide.

Methods and Results drafted from your own protocol and analysis, in a pre-structured manuscript. Nothing goes in until you accept or edit it.

Write-up — Results

Analysis

Spearman's ρ0.41
95% CI[0.21, 0.58]
p< .001
n84

Re-run on 84 responses

Draft · Results

AI-assisted · not inserted yet

Of 84 respondents, poorer sleep quality was moderately associated with higher perceived stress during the exam period (Spearman’s ρ = 0.41, 95% CI [0.21, 0.58], p < .001).

AcceptEditDrafted from your analysis — never invented.

Illustrative preview — fictional study, not real data.

Stage names and count differ by study type. See all 15 study types.

Standalone tools

Every tool also works on its own.

Need only a literature search or a quick analysis? Open the tool directly — no project required — and turn it into a project later if it grows into one.

  • Literature search

    PubMed and OpenAlex in one box.

  • Survey builder

    Branching, Likert, mobile-first link.

  • Statistical analysis

    Tests suggested from your variables.

  • Citation manager

    Vancouver, APA 7th, Harvard.

  • Manuscript editor

    IMRaD and CARE structures.

  • Qualitative codingSoon

    Highlight-to-code canvas.

Questions

Will this replace SPSS, Zotero, or Word?
No, and we don’t claim it will. No single Oshki tool tries to beat its specialised counterpart — Oshki connects a project’s tools into one chain, so you stop re-entering and reformatting the same work between them.
Do the stages change by study type?
Yes. Each study type has its own 4 or 5 stages and its own reporting guideline — a case report follows a shorter path than a cohort study.
Does the AI write my paper?
No. AI drafts from your own protocol and results, never invents data or citations, and nothing is inserted until you accept or edit it. Methodological decisions stay yours.
When does it launch?
We don’t have a fixed date yet. Join the waitlist and we’ll email you as soon as there’s something to try.
Will there be a free plan?
Yes, a free plan is planned alongside paid tiers for heavier use. Pricing is still being finalised.

Join the waitlist

Be first in when Oshki opens.

Two minutes: your role, the study you’re on or planning next, and the tools you use today. Optionally, a 15-minute call to tell us what’s missing.

  • — For medical & health-science students and researchers
  • — No launch date yet; we’ll email you when there is
  • — No spam. Unsubscribe or ask for deletion any time

Waitlist form not connected yet

Set TALLY_FORM_ID in lib/constants.ts to embed the Tally form here.