My take on "The AT-URI Syntax Mess": the canonical form should be percent-encoded, and today's colon syntax should stay as a defined lax input form. Longer writeup in the forum: discourse.atprotocol.community/t/the-at-uri-syntax-mess/934/…
The AT-URI Syntax Mess
discourse.atprotocol.community
bryan newbold
Jul 3, 2026 at 5:36 PM
Bringing back at://
I’d also like to voice a strong dissenting opinion: a separate URI scheme follows long-standing precedent, reduces cognitive overhead and software complexity by making the transport distinction explicit, and has less ambiguous error handling. I think this direction trades short-term convenience for long-term ambiguity. The transport models are incompatible, not variants of each other Public atproto records flow through a well-understood pipeline: repos export as CAR files, relays consume fireh...
The AT-URI Syntax Mess - at:// pizza thoughts
Why AT URIs are not valid under IETF RFC-3986, what the impact is, and some possible ways forward.
atstats - ATProto Lexicon Usage
Understanding Atproto - AT Protocol Docs - AT Protocol
Dive deeper into the concepts behind the AT Protocol.
Declare and resolve website links for at:// URIs. A demo of the me.byjp.atproto.deeplink lexicons.
Hey #atprotodev; should we make a convention for annotating HTML elements with the AT uris they’re built with? (So that, say, a browser extension could offer links to @pds.ls) Something like this? <li typeof="schema:Thing" resource="at://did:web:example.com/org.example.lexicon/rkey">
Seriously mild take: ATProto web clients should serve a <link rel="alternate"> containing an at:// URI I've just opened a PR to do this on Frontpage and atproto-browser. I'd for other clients to follow along with the convention! github.com/likeandscribe/frontpage/pull/…
Add AT URIs as alternate links by tom-sherman · Pull Request #172 · likeandscribe/frontpage
github.comzick is studiously thinking through the atproto+ucan combo here: discourse.atprotocol.community/t/musing-ucans-groups-communi… @expede.wtf (ucan editor & maintainer) chiming in 💖 @dholms.at also co-authored UCANs so his input here would be greatly appreciated!
Zicklag
I'm pretty sure UCANs are awesome. They're not the easiest thing to understand how to apply, and I need to see how they work out in practice, but I'm excited about the cool things that you can do with them. They've got some cool capabilities. 😉
If you're an #atprotodev I'd love your thoughts on this short proposal. I think we'd all benefit from a way to tag our HTML to reference the at:// uris they're derived from! @popfeed.social, @bookhive.buzz, @sifa.id, @leaflet.pub, @pckt.blog, @offprint.app and many more — penny for your thoughts?
Atproto markup for HTML
byjp.leaflet.pubI think this breaking change for standard.site needs to become an ecosystem-wide conversation and convention. Any web page that represents a corresponding AT record needs to include a meta tag that includes that record. <meta name=“at:uri” content=“at://…” \>
rich harris
atproto/@standard.site friends: any thoughts on this issue? This... <link rel="site.standard.publication" href="at://..."> ...is invalid HTML, per html.spec.whatwg.org/multipage/semantics.html#the-…, because DIDs aren't hostnames. This causes problems with some tools. Issue here: github.com/sveltejs/kit/issues/15935
zick is studiously thinking through the atproto+ucan combo here: discourse.atprotocol.community/t/musing-ucans-groups-communi… @expede.wtf (ucan editor & maintainer) chiming in 💖 @dholms.at also co-authored UCANs so his input here would be greatly appreciated!
Zicklag
I'm pretty sure UCANs are awesome. They're not the easiest thing to understand how to apply, and I need to see how they work out in practice, but I'm excited about the cool things that you can do with them. They've got some cool capabilities. 😉
HTTP gave us hypertext, atproto gives us "social hypertext"
Ronen Tamari
Needless to say, this is all very resonant with our work at @cosmik.network @semble.so - we’re passionate about developing hypertext systems that scaffold sensemaking and transformative learning journeys for individual and collectives. Excited to work on bridging theory and practice!
@tynanpurdy.com asked for this feature and I like it: every atproto app should support AT-URIs as top level routes, @pds.ls-style. Makes it super easy to integrate. So this would be an argument for e.g.: `https://bsky.app/at://did:plc:rbvrr34edl5ddpuwcubjiost/app.bsky.feed.post/3micfxauzu22c`
Support top-level AT-URI resolution · Issue #1012 · streamplace/streamplace
github.comOh, a way to share links and community blog posts. A great idea! We can share our homework 👀 discourse.atprotocol.community/t/moving-atprotocol-dev-off-o…
Moving atprotocol .dev off of Ghost
discourse.atprotocol.communityFinalizing our new Shortcuts system. In this example we're using @atmo.garden, a new integration. Not every AT Protocol service uses identifiers like c/, so users can now create their own shortcuts. For example, @semble.so could be s/, sem/, or even /👀. This allows fast and quick results 👀
