User:Linkgenetic

This is a Wikipedia user page.
This is not an encyclopedia article or the talk page for an encyclopedia article. If you find this page on any site other than Wikipedia, you are viewing a mirror site. Be aware that the page may be outdated and that the user whom this page is about may have no personal affiliation with any site other than Wikipedia. The original page is located at https://en.wikipedia.org/wiki/User:Linkgenetic.

About this account

Hello — I'm Linkgenetic, the Wikipedia account of Link Genetic, a Swiss research project working on persistent identifiers for the open web.

This user page exists for two reasons:

To introduce the LinkID project to the Wikipedia community.
To serve as a live demo of how easy it is to mint a persistent identifier for any external citation — without leaving the page you're already editing.

I am not here to promote a commercial product. The hosted resolver at linkid.io is free to use for public-interest citations, the API is open, and the project is being developed in the open.

What is LinkID?

LinkID is a persistent identifier ecosystem in the same family as DOI, Handle, and PURL. Given any public URL, it issues a stable, resolvable identifier of the form:

linkid:1f4a2b3c-...-9e8d7f6a5b4c   →   https://linkid.io/r/1f4a2b3c-...-9e8d7f6a5b4c

The resolver stores the original destination, the metadata extracted at mint time (title, description, language, OpenGraph / Dublin Core thumbnail), an archive snapshot reference, and a cryptographic mint receipt. When the upstream URL breaks, the resolver can fall back to an Internet Archive snapshot or a system-generated notice page rather than returning a 404.

Key properties relevant to Wikipedia citations:

IANA-registered linkid: URI scheme — citable as a first-class identifier alongside doi: and arXiv:.
Free public resolver at linkid.io running on a global CDN with sub-10 ms cache-hit resolution at 450+ edge locations.
Phishing protection — every URL is checked against Google Safe Browsing and OpenPhish before a LinkID is issued.
Deduplication — minting the same URL twice returns the same LinkID, so the same source can't accumulate orphaned identifiers.
Internet Archive integration — every mint can trigger a Save Page Now snapshot, giving the citation a recovery path the moment it's created.
GDPR Article 17 support — erasure requests produce a signed certificate; the resolver returns a clean 410 Gone instead of leaking deleted content.

The project is intentionally aligned with WP:CITE and the existing work on link rot and Wayback integration — it is meant to complement, not replace, those efforts.

Why I'm interested in Wikipedia

Wikipedia is the largest curated body of external citations on the web, and link rot is one of its largest long-term maintenance problems. InternetArchiveBot has shown that the community is willing to adopt automated tooling when it provably reduces dead links. LinkID is an attempt to add a complementary layer: instead of only repairing broken citations after the fact, mint a persistent identifier at citation time so the recovery metadata exists from day one.

I'm currently participating in:

Wikimedia Hackathon 2026 — Phabricator task T422252, integrating LinkID minting into the Wikipedia editing flow.
The IABot DOI Rot Pilot 2026 — a measurement project on broken DOI citations in Wikipedia references.
Wikidata alignment — exposing the linkid: namespace as a Wikidata external identifier property.

Live demo: mint a LinkID for a citation

Anyone can try this without an account. Paste any URL into the form at:

https://linkid.io/auto-mint

The page will:

Fetch metadata for the URL (via the MediaWiki REST API for Wikipedia-family URLs, or via OpenGraph / Dublin Core / <title> elsewhere).
Run the two-layer phishing check.
Return a linkid:<uuid> identifier plus a resolver URL of the form https://linkid.io/r/<uuid>.

The resolver URL can be dropped directly into a Wikipedia <ref> in place of a bare external link, e.g.:

<ref>"Example source". Retrieved 2026-05-22.</ref>

If the upstream source later breaks, the LinkID resolver still answers with the original metadata, an archive fallback, or a 410 erasure certificate — so the citation remains verifiable instead of silently rotting.

MCP: minting LinkIDs from an AI assistant

LinkID also ships a Model Context Protocol (MCP) JSON-RPC 2.0 server, so any MCP-compatible AI assistant — Claude Desktop, Cursor, ChatGPT desktop with MCP, and similar — can mint and resolve LinkIDs directly while drafting a citation. The relevant tools exposed are:

linkid.mint — given a URL, return a linkid:<uuid> identifier (runs the same phishing check and deduplication as the web UI).
linkid.resolve — given a LinkID, return the current destination, metadata, and archive fallback.
linkid.preview — read-only metadata extraction without minting.

In practice this means an editor who uses an AI writing assistant can ask it to "mint a LinkID for this source and insert a <ref>" and the assistant will return a stable identifier ready to drop into the article — no separate browser tab, no copy-paste. The MCP transport is documented at linkid.io/api-docs.

This is offered as a convenience for editors who already use such tools; nothing on Wikipedia requires it. The same identifiers can be minted from the web form at linkid.io/auto-mint, from curl, or from any HTTP client.

Editing this user page

Per WP:UP, this page is maintained by the Linkgenetic account. Other editors are welcome to leave messages on the talk page, or to make uncontroversial maintenance edits (typo fixes, dead-link tagging, category cleanup). For substantive changes please post on the talk page first.

External links

linkid.io — public resolver, documentation, and live demo
API documentation
Changelog (currently v5.6.0)
Link Genetic — the organization behind LinkID