NIP-B0: Web Bookmarking
NIP-B0 defines editable web bookmarks as kind 39701 events, making saved URLs portable, taggable, commentable and queryable by a normalized d tag instead of trapped inside one bookmarking app.
Bookmarks are personal infrastructure, not browser debris
People save links for later reading, research, memory, curation and sharing. Most bookmarking systems lock that behavior inside one browser or service. NIP-B0 makes a web bookmark a signed addressable Nostr event.
The event kind is 39701. The d tag is the URL without the scheme, while clients assume https:// or http://. That makes bookmarks queryable by URL and editable over time.
The content can hold a detailed description. Tags can hold title, topics and first-published time. Replies and comments use NIP-22, so discussion can attach to the bookmark without turning the bookmark itself into a normal note.
An addressable URL object with metadata
NIP-B0 does not try to replace all read-it-later features. It standardizes the object that such apps can share. A bookmark can have a title, tags, published time and a body description. The addressable key lets clients update the same bookmark rather than reposting duplicates.
Because the d tag removes the scheme, clients need to handle canonicalization carefully. The same web page can appear with tracking parameters, mobile domains, trailing slashes or different schemes. Good bookmarking clients should normalize without destroying meaning.
The standard is useful for link collections, public research libraries, personal archives and community-curated lists.
The NIP grew from practical bookmark-app needs
Sebastian Hagens added more requirements for kind 39701 web bookmarks in April 2025 through PR #1849. fiatjaf uppercased B0 and B7 the same day, and AsaiToshiya cleaned up B0 in August 2025.
External mentions point to a small but real ecosystem. Key Pair discussed bookmarking on Nostr and referenced Pinstr and Yumyu. Awesome Nostr lists Community Curated Nostr Stuff around NIP-B0-style link aggregation.
The best reading is that NIP-B0 gives bookmarking apps a shared spine. The interesting product work is search, reading state, tags, export, comments and trust.
A bookmark client should care about canonical URLs
A serious client should normalize tracking parameters, preserve the user's chosen title, allow private or public curation choices, show comments through NIP-22 and keep the saved URL easy to export.
Tags should support discovery without becoming spam bait. Public bookmark feeds are only valuable when they are curated enough to trust.
The most useful interface is not a feed of links. It is a personal or community library where URLs can be found again.
The hard problem is duplicate and low-quality links
A bookmark standard can fill with scraped links, tracking URLs and duplicates if clients do not normalize and curate. Protocol support is only the first layer.
Public bookmarks can also reveal research interests or private plans. Apps should not assume every bookmark belongs in public.
Direct sources
Use the official file first, then the commit history, implementation references and adjacent standards. NIPs move, and product guidance gets weaker when those source trails are hidden.
