Research Agents

Nubint's AI research agent is one unified chat that handles everything — topic brainstorming, drafting, review. It calls dedicated workflows automatically: Literature Review, Peer Reviewer, Proofreader. AI Paper Search adds semantic discovery across 460M+ papers.

Open AI Chat and pick an agent from the list above the chat input, or click an agent in the left sidebar. Each run deducts credits based on your plan, and results are saved to your chat history, library, and documents for later reference.

The Hypothesis Generator turns a vague research question into 3 testable hypothesis candidates, each grounded in 460M+ papers and scored on testability, novelty, feasibility, and clarity. Pick the strongest one and refine it to fit your study direction.

The Hypothesis Evaluator scores any hypothesis on 4 academic criteria — testability, novelty, feasibility, clarity — on a 0–100 scale. Novelty is measured against 460M+ live papers, and weak criteria come with concrete fixes, not vague feedback.

The Survey Generator takes a hypothesis and auto-generates 10–30 structured survey items for validation — Likert, multiple choice, and open-ended. Items are built on validated measurement scales from 460M+ papers, with reliability info included, and export to DOCX, CSV, or JSON for Google Forms, Qualtrics, and SurveyMonkey.

The Literature Review agent collects related papers and analyzes up to 40 at once, organizing findings, methodologies, and trends into a structured report with auto-generated references.

The Research Tracer starts from one seed paper and bidirectionally traces its citation network — backward to seminal references and forward to follow-up work — up to 5 hops. The output is a visualized lineage with AI-identified inflection points that explain why a paper shifted the field.

The Citation Recommender takes a claim or hypothesis and surfaces the 3–5 strongest supporting citations from 460M+ papers using semantic search. Each recommendation comes with the exact supporting quote, page info, and citations auto-formatted in BibTeX, APA, MLA, and Chicago.

The Figure Generator turns a plain-language description into a publication-quality scientific figure in seconds. Describe an experiment pipeline, system architecture, or conceptual schematic and it draws a clean, accurately labeled figure on a white background — no watermarks, no baked-in captions — ready to download as PNG or JPG for your paper or slides.

The stop button in AI Chat immediately halts any ongoing response, while toggles let you control external and library paper search independently. Keep external search on during exploration and switch to library-only during writing to focus on papers you already have.

The Add Context button lets you select specific papers or collections from your library so AI bases its responses on those materials. Use it when asking about a specific paper, analyzing a collection, or specifying references for AI editing. Contexts appear as removable chips.

Click the New Chat button at the top of the chat panel to start a fresh conversation while auto-saving the previous one. Use it when switching research questions or when a conversation loses focus, and return to previous chats anytime from the chat list.