Sharing what an AI tool builds — Claude artifacts, vibe-coded apps, Codex bundles — before you deploy.
Start here
From a Claude artifact to a vibe-coded app to a multi-file Codex bundle — how to share what an AI tool builds, as a private link, without a deploy pipeline.
You sent a Claude Slides deck with 'anyone with the link' and your client got a sign-up screen. Why that happens, what Docs, Slides and Design actually export, and how to get the work in front of someone who will never make a Claude account.
You sent a client the public link to a Claude Code artifact and asked them to mark up what to change. There's nowhere to comment. That's by design. Here's the rule, which plans it leaves without a comment option for outside viewers, and how to collect pinned notes anyway. Checked against Anthropic's docs on 15 September 2026.
Hit a ChatGPT Sites usage limit? Your Sites aren't gone. Here's how to tell the ones that are just pages, which travel, from the ones that need Sites underneath them.
The prototype took an afternoon. Getting it in front of the person who has to say yes takes longer — because a repo-and-deploy is too much ceremony, an attachment opens as code, and a platform link drags your reader onto the platform. What the handoff actually needs.
There's a third option between a free public download and a locked paywall: let people earn the freebie by engaging — a comment, a reply, a DM keyword. A private, unguessable link is the delivery layer that makes the engagement-gate work without an account wall or a public file.
An AI tool writes you a working mini-tool — a calculator, a quiz, an estimator — from one prompt. Put it on a private link you control and it becomes a stronger lead magnet than any PDF, because people use it instead of just downloading it. Here's the conversion loop, and where the browser ends and a backend begins.
The comment-to-DM play, end to end. You reply with a freebie, move the conversation to a DM, and let one private hosted link carry the handoff. Here's the motion step by step — what to say, where the link goes, and why a private hosted page beats an attachment or a public deploy.
The permission prompt arrives mid-session and the honest answer is: by default, only you. What publishing actually does, when a link becomes public, and where Anthropic's own pages disagree — on which plans have it, and on whether a public link needs a Claude sign-in. Checked against their docs on 12 September 2026, sign-in rechecked 19 September 2026.
An agent writes a report, tells you the file path, and hours later the download button does nothing. The output was living on the computer that made it, and that computer was designed to disappear. Why this failure is structural rather than a bug, and what to do at the moment the work is finished.
Codex can now share a thread as a read-only link. It shows someone how the work was reasoned — not the working thing it produced, and not a copy that keeps up with you. Here is exactly what the snapshot carries, the four boundaries the announcement doesn't spell out, and how to tell which of the two jobs you're actually doing.
The AI tool built the page. Sending it to a client is a second job: stripping the construction site, making it read as yours, and putting it behind a private link. A practical checklist for turning AI output into a client-ready handoff.
The Gemini share button fails and the error tells you nothing. Usually it isn't a bug and it isn't your file — public link sharing is off by default on work and school accounts, and the switch is one only an admin can flip. Here's how to tell, and what to do when you can't wait for them.
A disabled Publish button isn't one bug — it's four different situations wearing the same grey. Two of them are permanent, two clear in a minute. Here's how to tell which one you're looking at, and how to get the page to the person waiting for it either way.
HTML has quietly become a document teams work on together — reports, decks, small tools. But a link is a reading surface, and the thing your colleague's AI needs to open is the source. Where each one lives, and why collapsing them causes the mess.
AI builders often export an app as a folder — index.html plus separate CSS, JS, and assets — not a single file. How to share the whole thing as one private link without flattening it or deploying it.
You built an app in Bolt, v0, Replit, Lovable, or Claude and it works — but it only works on your machine, and the deploy pipeline is a bigger commitment than the question you're trying to answer. What fits the in-between moment: a live link in seconds, and you decide who opens it.
A lead magnet on a link that self-destructs in days is a slow leak in your funnel. Why marketers and coaches need a never-expire private link, and how one URL turns a freebie into an evergreen, semi-private channel from comment to DM.
The message is about Anthropic's side, not yours, and it usually clears on its own. The risk isn't the outage — it's what people press while waiting. Here's the safe order of operations, and how to get the work out of the artifact while you wait.
You built a page with an AI tool. Three places it can live — the builder's own share URL, a public deploy, or a private link — and a clear way to decide between them based on whether the page has a backend and who is supposed to reach it.
You built something in Claude, came back weeks later, and it's gone. Usually it isn't deleted — it's sitting inside a conversation you can no longer find. Here's how to get it back, and the one case where it really is unrecoverable.
Artifacts handle more than people expect right up until they don't. The four limits that actually stop projects — no backend, storage that only exists after publishing, an irreversible unpublish, and the settings toggle — and what to do at each one.
Artifacts stop appearing and every guide says the same four generic fixes. Here's the order that actually separates a setting on your side from an outage on theirs — starting with the one documented cause that produces total failure.
The artifact works in the chat, publishing succeeds, and the link opens to nothing. Three different failures produce that same white screen, and one thirty-second check tells you which one you have.
It looked like a URL. You sent it. Most of the time it never existed — but one of these is the opposite case: <private_url> means an address was there and got redacted. How to tell which one you have, in one second, and what each one actually needs.
The agent handed you a URL and said it's live. It probably is. What it didn't tell you is who can open it, how long it lasts, whether the address is guessable, and what happens to it when you revise — four defaults chosen for you, none of them announced.
ChatGPT Sites can now invite a named viewer from outside your workspace — but they have to sign in with the account you invited. If your client won't, here's the shape of what's left and the route round it.
The steps take about ten seconds. The cost is less obvious: unpublishing permanently retires that artifact's ability to be published, and deletes its stored data. Here's how to do it, what it takes with it, and what to reach for when access only needs to end for now.
The normal delivery ritual is finish, then send — and every revision mints a new URL and another 'ignore my last email'. If your AI tool can publish and then update in place, the link can go out first and stay correct. Including when that's a bad idea.
Claude refuses to share your artifact and suggests publishing a new version — which often doesn't clear it. Here's the diagnosis that actually separates the two failures behind that message, and the export route that doesn't depend on the toggle working.
Not to a public link — on any plan. On Pro and Max, a connector-backed artifact can't be shared at all. Here's the mechanism that makes that rule inevitable, and what to do once you know which page you're actually building.
The iteration loop with Claude / Cursor / Lovable produces a new version every round. Channels that let each round travel cleanly — without exposing the chat, the previous version, or the next reviewer to noise.
A step-by-step: turn on Review for a private link, let anyone pin comments to the exact spot (no account), then export the thread as a prompt to paste into Claude or Codex.
You shared the AI-built page — then what? The real work is the loop: build, show, react, fold the reaction back into the next prompt. Why hosting is only step one, and what actually closes the loop.
"It asks me to log in." "It's blank." The link isn't broken — it was never built to leave the tool. What each AI tool actually hands you, and what opens.
An AI tool builds the landing in fifteen minutes; deciding where it lives takes the next hour. Three hosting shapes and how to pick between them.
Sending a Claude or ChatGPT artifact to one client — how the channels compare, where private-link hosting fits, and the six clicks that get you there.