# Explore Science > An AI co-scientist for working researchers. Explore Science reviews manuscripts, chats about a paper it has actually read, searches the literature, and checks novelty, all from the same engine the lab uses to publish its own autonomous research. Explore Science is an AI science laboratory. Every output, the manuscript review, Rosa the co-scientist chat, the novelty and integrity checks, runs on one engine that mixes multiple frontier models with an in-house science stack. A review takes about an hour, often more, because rigour takes time; every paper receives a transparent 0-100 score that combines review quality with the field-specific evidential weight of the study design, with the full methodology and per-field evidence hierarchies published at https://explorescience.ai/scoring. ## Core pages - [Explore Science](https://explorescience.ai): What the platform is and who it is for: an AI co-scientist for working researchers. - [How it works](https://explorescience.ai/how-it-works): The autonomous review pipeline, built on a mixture of models and an in-house science stack. - [How we score a paper](https://explorescience.ai/scoring): The 0-100 scoring methodology in full: review quality combined with the evidential weight of the study design. - [Features](https://explorescience.ai/features): Everything the platform does: manuscript review, Rosa the co-scientist chat, projects, the Calibre score, editorial review, novelty check, and teams. - [Rosa, the co-scientist chat](https://explorescience.ai/chat): Rosa reads your manuscript, verifies its references, and answers from real evidence. - [Research](https://explorescience.ai/research): Research from the lab that builds and publishes its own autonomous science. - [Pricing](https://explorescience.ai/pricing): Free, Researcher, Pro, and Institution plans. The same engine on every tier; only the volume changes. - [FAQ](https://explorescience.ai/faq): Questions scientists ask: model mixture, training, peer-review comparison, field coverage, and institutional policy. ## Integrations - [Integrations](https://explorescience.ai/integrations): Explore Science integrations: save sources to Zotero, sign in with ORCID, send a preprint from bioRxiv or medRxiv, and review from your MCP client. - [Zotero integration](https://explorescience.ai/integrations/zotero): Connect your Zotero library to Explore Science. Save the sources from a review, from Rosa and from a novelty check to Zotero, filed by project. - [ORCID integration](https://explorescience.ai/integrations/orcid): Sign in to Explore Science with your ORCID iD. Your public ORCID record verifies your institution and links your published work. - [openRxiv external author service for preprint review](https://explorescience.ai/integrations/openrxiv): Explore Science is an external author service on bioRxiv and medRxiv. One click in the Author Area sends your preprint for review. The first is complimentary. - [AI review of a bioRxiv preprint, from your Author Area](https://explorescience.ai/integrations/biorxiv): One click in your bioRxiv Author Area sends the preprint to Explore Science for a full review, each issue tied to its passage. The first is complimentary. - [CONSORT and STROBE checks for your medRxiv preprint](https://explorescience.ai/integrations/medrxiv): One click in your medRxiv Author Area sends the preprint to Explore Science for a CONSORT or STROBE check, marked against your text. The first is complimentary. - [Explore Science MCP server](https://explorescience.ai/mcp): Connect Explore Science to your MCP client. Send a manuscript for review and work through its Calibre score and findings in your chat. ## Reporting guidelines - [Reporting guidelines](https://explorescience.ai/reporting-guidelines): Which reporting guideline governs a study, and a check of a manuscript against it. - [CONSORT 2025 checklist](https://explorescience.ai/reporting-guidelines/consort): Consolidated Standards of Reporting Trials, every item in full, for a randomised trial. - [SPIRIT 2025 checklist](https://explorescience.ai/reporting-guidelines/spirit): Standard Protocol Items: Recommendations for Interventional Trials, every item in full, for a trial protocol. - [STROBE 2007 checklist](https://explorescience.ai/reporting-guidelines/strobe): Strengthening the Reporting of Observational Studies in Epidemiology, every item in full, for an observational study. - [PRISMA 2020 checklist](https://explorescience.ai/reporting-guidelines/prisma): Preferred Reporting Items for Systematic Reviews and Meta-Analyses, every item in full, for a systematic review or meta-analysis. - [ARRIVE 2.0 (2020) checklist](https://explorescience.ai/reporting-guidelines/arrive): Animal Research: Reporting of In Vivo Experiments, every item in full, for an animal study. - [CARE 2013 checklist](https://explorescience.ai/reporting-guidelines/care): CAse REport guidelines, every item in full, for a case report. - [SQUIRE 2.0 (2015) checklist](https://explorescience.ai/reporting-guidelines/squire): Standards for Quality Improvement Reporting Excellence, every item in full, for a quality improvement study. - [CHEERS 2022 checklist](https://explorescience.ai/reporting-guidelines/cheers): Consolidated Health Economic Evaluation Reporting Standards, every item in full, for an economic evaluation. - [AGREE 2016 checklist](https://explorescience.ai/reporting-guidelines/agree): AGREE Reporting Checklist, every item in full, for a clinical practice guideline. ## Blog - [What an AI review finds in a bioRxiv preprint](https://explorescience.ai/blog/what-a-deep-review-finds-in-a-biorxiv-preprint): Know what reviewers will say before you submit your bioRxiv preprint to a journal. - [How to check a medRxiv preprint against CONSORT or STROBE](https://explorescience.ai/blog/check-a-medrxiv-preprint-against-its-reporting-guideline): Check your medRxiv preprint against CONSORT, STROBE or the guideline your study follows, before you submit it to a journal. - [Get a review of your bioRxiv or medRxiv preprint before journal submission](https://explorescience.ai/blog/explore-science-on-biorxiv-and-medrxiv): Explore Science is now an external author service on bioRxiv and medRxiv: a review or guideline check from your Author Area. The first is complimentary. - [AI tools for manuscript review in 2026](https://explorescience.ai/blog/ai-tools-for-manuscript-review): We compare fifteen AI tools that review a manuscript before submission: what each reads, how it scores, its reference checks and price. Checked 15 October 2026. - [Welcome to the new home of Paper-Wizard: Explore Science](https://explorescience.ai/blog/paper-wizard-is-now-explore-science): Paper-Wizard has grown into something larger. Here is what that means: - [How to do a good peer review?](https://explorescience.ai/blog/how-to-do-peer-review): Learn how to do a constructive, well-structured and useful peer review. - [How pre-registrations, registered reports and well written research proposals improve your research](https://explorescience.ai/blog/pre-registration-research-planning): Learn how pre-registration and registered reports can improve your research quality and help combat questionable research practices. - [Embodied AI Scientists: Running Experiments in the Physical World](https://explorescience.ai/blog/embodied-ai-scientist): The first AI system to autonomously conduct end-to-end research with human participants, as AI steps out of the computational realm into the physical world. ## Research Papers written end-to-end by the autonomous system, published under the lab's own DOIs. - [Rigorous model comparison reveals linear speed-reasoning coupling](https://explorescience.ai/research/rigorous-model-comparison-reveals-linear-coupling): Autonomously generated by Explore Science. - [Battery diversity is associated with simple structures in neuropsychological assessment](https://explorescience.ai/research/battery-diversity-simple-structures-neuropsychological): Autonomously generated by Explore Science, DOI 10.63708/exsci.6. - [Trail Making complexity negatively associates with reasoning beyond processing speed](https://explorescience.ai/research/trail-making-complexity-negatively-associates-reasoning): Autonomously generated by Explore Science, DOI 10.63708/exsci.5. - [Imagery vividness fails to predict serial dependence in visual working memory and mental rotation](https://explorescience.ai/research/imagery-vividness-serial-dependence-visual-working-memory): Autonomously generated by Explore Science. - [Independence of visual working memory precision and mental rotation performance: theoretical and methodological implications](https://explorescience.ai/research/independence-visual-working-memory-mental-rotation): Autonomously generated by Explore Science. - [Visual memory precision shows negligible spatial task links](https://explorescience.ai/research/visual-memory-precision-spatial-task-links): Autonomously generated by Explore Science. - [Strategic visual memory allocation during complex mental rotation: Evidence for adaptive cognitive resources](https://explorescience.ai/research/strategic-visual-memory-allocation): Autonomously generated by Explore Science. ## Optional - [Trust](https://explorescience.ai/trust): Security and compliance posture. - [Data handling](https://explorescience.ai/data-handling): How manuscripts and research data are stored and used. - [Privacy policy](https://explorescience.ai/privacy): How personal data is handled. - [Terms](https://explorescience.ai/terms): Terms of service.