Use cases

One research agent across the whole workflow — from the first literature search to the submitted manuscript. Every artifact below carries the code and data that produced it, so your study is reproducible from day one.

STEP 01

Literature

Review the field with every claim cited

Ask a question and the agent fans out across PubMed, Crossref, OpenAlex, Semantic Scholar, and more — plus Scopus, Web of Science, and KCI when connected — then composes a report where every sentence carries an [n] citation to a real source. Uncited claims are removed, not published.

Cited review reportSource tableVenue-tier breakdown
STEP 02

Hypothesis

Turn findings into testable hypotheses

From the reviewed literature, the agent helps you articulate constructs, expected directions, and competing explanations — and drafts a hypothesis map that links each prediction back to the studies that motivated it.

Hypothesis mapConstruct definitionsRival-explanation notes
STEP 03

Experiment

Design studies with power you can defend

Describe the study you want to run and the agent surveys how related studies designed theirs, then produces a structured design document — protocol flow, measures, sample plan with a power curve — ready to hand to Qualtrics, PsychoPy, or your IRB.

Design documentProtocol flow diagramPower curve
STEP 04

Data Analysis

From raw signals to analysis-ready data

Load survey exports, interaction logs, or biosignals (facial AU, gaze, EDA, HRV) once per session and work by reference afterwards. Cleaning, merging, and exploratory checks surface missing data and anomalies explicitly instead of silently dropping them.

Cleaning logEDA · HRV time seriesGaze heatmap · AOI
STEP 05

Statistics

Modern models with assumptions checked twice

Run factor analysis, SEM, reliability, mixed-effects, and Bayesian models through conversation. A separate review agent checks assumption violations, multiple comparisons, and sample-size concerns before results reach your report.

SEM path diagramFactor-loading heatmapReliability α · ω table
STEP 06

Writing

Manuscripts that never drift from the data

Methods and results sections come pre-filled from the analyses you actually ran, with every figure and table linked to its code and data source — so revisions reproduce in one click and nothing in the paper drifts from the evidence.

Manuscript draftPublication-ready figuresReproducibility bundle

Curious how these run in practice? See the skills and connectors behind each step.