Link
Reusers must provide direct access to the original Wikimedia project page(s) from which the reused content was drawn (e.g., the specific Wikipedia article, the Commons file page, or the Wikidata entity).
This is an essential attribution requirement that preserves verifiability by letting end users access information at the source with a single click. Links can also contribute to sustaining open knowledge by returning traffic to the Wikimedia ecosystem, where readers can verify information, learn further, or even contribute edits.
Links can take the form of text (e.g., the source project's name, the title of the article or image being shared), be represented by a complete URL, or be shaped as affordances (e.g., “Read full article”).
Data sources
This section identifies where and how the data behind this signal can be obtained. It points to the relevant Wikimedia APIs, datasets, or metadata fields that reusers can rely on to implement the signal accurately and consistently.
Wikimedia Attribution API Beta Recommended
The Wikimedia Attribution API returns a canonical link to the Wikimedia page being reused, including Wikipedia articles and Wikimedia Commons file pages.
Reference documentation can be found in any wiki's REST API sandbox, such as the REST API sandbox on English Wikipedia →
Share feedback on this beta API →
Wikimedia Enterprise API Recommended
Unlike the MediaWiki Action API and the Wikimedia REST API, specifying the language code and project isn't necessary to retrieve URL information with the Wikimedia Enterprise On-demand API. Instead, it's possible to simply make a call to retrieve all articles from all Wikimedia projects and language versions that match an article's name. You can use fields and filters to restrict your search to the languages and Wikimedia projects you're interested in.
See the Wikimedia Enterprise API documentation →
MediaWiki Action API
If a page title is available, the query endpoint can be used to retrieve basic metadata about the page, including the last-touch data, and links for the page itself.
See the MediaWiki Action API documentation →
URL manipulation
URLs can be constructed based on the known language, project, and title. In general, Wikimedia projects’ URLs follow the pattern of: https://{language}.{project}.org/wiki/{title}.
This structure includes Wikimedia Commons, when referencing the file by title. For example: https://commons.wikimedia.org/wiki/File:Grasshoppper_(27703165173).jpg
Implementation guidance
Minimum requirements
Allow users to access the exact Wikimedia project page being reused; never link only to the project's homepage (e.g., wikipedia.org). For example, link to article pages on Wikipedia, to the file description pages in Wikimedia Commons, or entity pages in Wikidata.
Every source requires a link. When content is drawn from multiple pages, allow users access back to each individual source of information.
Always make sure that users understand how to access the original source of the information being presented. Avoid exposing raw URLs only, and opt for concise, descriptive labels instead. You may use the Source or Title signals as text for the link, or provide a clear navigation affordance that fits your context and is clearly associated with the reused content.
Best practices
In visual interfaces, you may also expose full links or breadcrumbs to the source next to the reused content as a way of revealing the navigation path to users.
In constrained environments, like small screens or compact UIs, you may use icons, short links, or other affordances to ensure one-tap access back to the source.
Reuse scenarios
Together with the Source signal, providing clear access to the original material being reused stands as the most essential attribution signal. Providing links to original materials is required in all digital, text-based contexts:
Search
Required
A direct link enables users to verify results and continue exploring the original source to further satisfy their information needs.
AI assistants
Required
Linking back to the specific article offers transparency and supports explainability in AI-generated responses.