@linkoi/coreLibraryA URL in. Metadata out.
Extract from a URL or HTML you already have. Get a consistent result, with Metascraper filling gaps when a page needs a closer look.
View the code Meet Linkoi Developer preview
A little more behind every link.
Extract titles, images, and context from a URL.
Give the links in your app something more to say.
TypeScript library. Your infrastructure.
From a URL to useful dataURLhttps://kitmo.app
MetadataNormalized output
kitmoMade by you.
Seen everywhere.
Seen everywhere.
One consistent response application/json
resolve(url)One URL. A structured response. Ready for your app.Read the page. Keep the context.
Open GraphTwitter CardsJSON-LDHTML
A small library.
Room to build.
Add metadata to bookmarks, link previews, and content tools. Choose the extractor or bring your own HTTP endpoint.
@linkoi/workerCloudflareYour endpoint. Your rules.
Deploy a metadata API with KV caching, background refresh, and optional bearer authentication. Run it in your Cloudflare account.
See the extraction APIMake a URL
worth opening.
Use the same metadata in different parts of your app. You decide how it looks and where it goes.
Bookmarks with context.
Save a title, description, and image alongside the URL. Make a collection people can scan.
Previews that say more.
Turn pasted links into useful cards in chats, notes, and editors. Style every field to fit your product.
Content you can organise.
Add publishers, authors, dates, and language when available. Give imported links more useful fields.
Illustrative interfaces built with metadata. Linkoi provides the data, not these UI components.
Built for developersLess parsing.
Less parsing.
More building.
Start with a URL. Use the fields you need.
The same API works across your server-side tools.
Developer preview. Packages aren't on npm yet.
Build the local workspace to try the API.
Under the surface.
Pages describe themselves in different ways.
Linkoi brings those details into one result.
Read the source
Fetch the page's HTML, or pass in HTML you already have. Configure fetch timeouts and size limits.
URL or HTMLFind the metadata
Read Open Graph, Twitter Cards, JSON-LD, and HTML tags. Use the optional Metascraper fallback for sparse results.
Extract + normaliseBuild with the result
Use a consistent set of fields in your app. The Worker adds a cache and background refresh for HTTP requests.
Structured metadataFocused on metadata.
Titles, descriptions, images, favicons, publishers, authors, dates, and languages. Optional YouTube API support when you need it.
Know what you're getting.
Linkoi reads the HTML a server returns. Browser rendering, crawling, and screenshots aren't included.
Source-available under PolyForm Shield. Non-competing commercial use is permitted.
A few things
before you dive in.
What's included, where it runs,
and how you can use it.
Is Linkoi a hosted API?
A public preview endpoint is available to try. For your own deployment, use @linkoi/core in your server code, or deploy @linkoi/worker to your own Cloudflare account.
Can I use it in a commercial app?
The current PolyForm Shield license permits non-competing commercial use. It restricts competing products and services, including free competitors. Linkoi is source-available, not OSI open-source. Review the repository's license before adopting it.
Does it render JavaScript or take screenshots?
No. Linkoi reads the HTML returned by the server. Metadata added only after JavaScript runs may not be available. Browser rendering, screenshots, and crawling aren't included.
Will every page return every field?
No. Results depend on what the page provides and whether it can be fetched. Handle missing fields in your UI. The optional Metascraper fallback can fill some gaps.
Can I try it today?
The packages are in developer preview and haven't been published to npm. Try the HTTP API or build the local workspace. The examples on this page use static data and don't make extraction requests.
