SciSpace Academy
Tutorial

Complete Guide to SciSpace Agent: Use Cases, Demos and Best Practices

SciSpace AcademyEnglishStart here

TL;DR

1-min read

This hands-on tutorial tours what SciSpace Agent can do across the research lifecycle, following real workflows from an initial question to finished outputs. It is designed for researchers, PhD students, postdocs, principal investigators, librarians and research administrators who want an end-to-end view of the agent and practical ideas for their own projects.

Key takeaways

  • Literature discovery combines semantic and full-text search with filtering, deduplication and evidence mapping, which charts where research on a topic is concentrated and where it is thin.
  • Paper analysis includes chatting with PDFs, extracting methods and claims, synthesising across studies and working with citation networks to understand how findings connect.
  • On the writing side, outlines can be turned into cited drafts, text can be paraphrased responsibly, and documents can be converted between Markdown, LaTeX and DOCX.
  • Data work covers extracting tables, cleaning and analysing data, building publication-quality charts and summarising results within the same environment.
  • A grants segment shows how to scaffold specific aims, gap and impact narratives, risk mitigation and compliance sections for a proposal.
  • The Prompt Builder and Agent Gallery offer ready starting points, and throughout, the agent is treated as an assistant whose sources and interpretations the researcher checks.
Apply it

To apply this, pick one stage of your current project, start from the Prompt Builder or Agent Gallery in SciSpace Agent, and verify the sources behind every output.

How to do it

  1. Start a task with Prompt Builder

    Open SciSpace Agent and use the Prompt Builder or Agent Gallery to start a new task quickly. This SciSpace Agent tutorial step saves you writing prompts from scratch.

  2. Discover and map the literature

    Run semantic and full-text searches, filter and deduplicate results, then build an evidence map. This gives a clean, organised evidence base.

  3. Analyse papers and extract claims

    Chat with PDFs to extract methods and claims and synthesise across studies. This turns reading into structured insight.

  4. Turn outlines into cited drafts

    Convert an outline into a cited draft, paraphrase where needed and export to Markdown, LaTeX or DOCX. This moves you from notes to a usable manuscript.

  5. Analyse data and create charts

    Extract tables, analyse the data and create publication-quality charts. Visuals strengthen your results.

  6. Scaffold grant narratives

    Use the agent to scaffold specific aims, gap and impact narratives, and risk mitigation for a grant. This gives your proposal a solid structure.

Watch the tutorial36 min

SciSpace Agent Guide: Use Cases, Demos and Research Best PracticesPlay video: SciSpace Agent Guide: Use Cases, Demos and Research Best Practices

Chapters

  1. Welcome and agenda: six SciSpace Agent use cases
  2. What is the difference between an AI research tool and an AI agent?
  3. How to run multiple research tasks in one AI prompt
  4. How to use SciSpace Agent Prompt Builder and Agent Gallery
  5. How do SciSpace Agent credits work?
  6. SciSpace Agent interface: live activity, outputs, CSV export and sharing
  7. How to find relevant research papers across Google Scholar, PubMed and arXiv at once
  8. Can an AI agent search ScienceDirect or any website for papers?
  9. Standard search vs Deep Research in SciSpace Agent for literature review
  10. What does AI Deep Research actually do? Five rewritten queries in parallel
  11. How to find research gaps with an AI agent
  12. Can AI write a research report in LaTeX? SciSpace Agent demo
  13. How to use AI for data analysis: upload a CSV and get charts and insights
  14. Can an AI agent build and publish a website for your research or conference?
  15. How to design a research poster with AI
  16. Closing: try your own SciSpace Agent queries

Try this on SciSpace

Frequently asked questions

What is SciSpace Agent used for?

SciSpace Agent carries out multi-step research tasks such as finding and screening literature, analysing papers, drafting text with citations, working with data and charts, and outlining grant sections. It is best treated as an assistant that speeds up groundwork, while the researcher checks sources, interprets results and makes the final decisions.

What is evidence mapping in a literature review?

Evidence mapping is a way of charting what research exists on a topic, often by study type, population, intervention or outcome, to show where evidence is concentrated and where it is thin. It helps identify gaps and plan a focused review, but it does not by itself assess the quality of individual studies.

Related tutorials