OSCR

Terms of use

Using OSCR means accepting these terms. They are short because the service is simple: a free registry, run on free plans, that asks little of you and promises only what it does.

The service

OSCR is a free registry of the code published with open-access neuroscience papers: records of papers and their code, copies of scripts where licences allow, a reader that shows a paper beside its code, tracing maps, and the means for authors and others to correct, validate and request removals. There is no fee, no paid tier and no advertising.

Who provides it

The service is provided by its sole operator, whose name and postal address are published here and on the privacy page before the registry's public launch.

Using the site

  • Reading, searching and downloading need no account and no acceptance beyond these terms.
  • Automated access is welcome at a reasonable pace: the site runs on a free plan whose daily requests are shared by every reader. For bulk use, take the open datasets rather than crawling pages.
  • Do not try to break into the service, to get around its limits or its security, or to make it publish what its rules forbid. Report a vulnerability privately through GitHub's security advisories.

Accounts

  • You sign in with an ORCID, GitHub or Google identity that is yours. OSCR asks for no email address and no password.
  • A role you hold — verified author of a paper, maintainer of a repository — lets you act for that paper or repository: use it only for them, and only truthfully.
  • The operator may refuse a request, withdraw a role or close an account that is misused: false claims, impersonation, abuse of requests, or conduct against the code of conduct.

What you contribute

  • Links you submit or correct become part of a paper's public record, as facts about where its code and data are. Who made a correction is not published: the record says "a verified author" or "a maintainer".
  • Your notes, statements and justifications are read by the operator and are not published.
  • A validation puts your name and ORCID iD on the map's public Zenodo record, as one of its creators, permanently.
  • You give the operator the right to use what you contribute to run the registry, and you confirm that it is accurate to your knowledge.

The content and its licences

  • The authors' scripts keep their repositories' licences; reuse them as those licences say (the code policy).
  • Abstracts and availability statements, shown only under CC BY, CC0, CC BY-SA or CC BY-NC, keep the paper's licence (the licence policy).
  • The records' metadata, when released as open data, and the tracing maps are under CC0-1.0.
  • The software is under the Apache-2.0 licence (its source).
  • The name and the logo identify the registry; use them as the brand page says.

Accuracy, and the source

Records are made by a harvester, and matches by a method, not by people: they can be wrong or incomplete, and every page says how each verdict was reached. The paper and the repository are the reference: when OSCR and the source disagree, the source prevails. A copy is a convenience, not an archive of record; for that, see Software Heritage, which the pages indicate when it has archived a repository.

Availability

The service runs on free plans and the operator's own computer, with no guarantee of availability. Parts of it stop for the day when a free quota is spent (the search says so), and features may change or end. The static pages, the datasets on Hugging Face and the maps on Zenodo do not depend on the rest.

Liability

The service and its content are provided as they are, free of charge, without warranty of any kind. To the extent the law allows, the operator is not liable for any loss arising from their use, including the use of code shown or linked here, which remains its authors' and is governed by its licence. Nothing in these terms limits a right you have under the law that applies to you, including your data protection rights (privacy).

Changes

These terms may change with the service; their history is public in the project's source. Last updated: 29 September 2026.