Planning workspace
Company priorities
A shared plan for the next chapter.
Org-managed AI artifacts
Turn AI-made files into secure, shareable sites.
Connect your agents and publish documents and sites to a company-owned link—with access controls built in.
acme.every.link/priorities
Customer synthesis
What we heard this quarter
Patterns from 42 customer conversations.
Positive signal+24%
Component library
Interface building blocks
Live performance
EOY Forecast
Revenue, retention, and momentum in one place.
One consistent handoff
Whatever AI model your team uses, the result is an org-owned link with granular access built in.
acme.every.link/2026-marketing-insightsWorks with
- Claude
- OpenAI
- Microsoft Copilot
- Gemini
- and more
How it works
Your agent, publishing in three steps.
MCP is the direct path for web AI. Coding agents use the official skill and CLI. Drag and drop remains available for a manual one-off.
- 01
Claim your link
Create
your-org.every.link—a company-owned home for everything your team publishes. - 02
Connect your agent
Add the remote MCP server to a web AI tool, or install the official skill and CLI in a coding agent.
- 03
Publish securely
Ask your agent to publish. Each member connects as themselves, and every deploy inherits the site URL and access rules.
Access
You choose who can open it.
Set it per site, change it later, and the URL never moves.
Public
Anyone with the link. Turn on indexing when you want it found.
Password
One shared password, no account needed.
Default
Sign-in required
Named people, or anyone on your email domain.
GoogleAppleGitHubMicrosoftSAML on Business
Backend, optional
A database without writing a backend.
Create a collection and its schema in the dashboard. Your site gets REST endpoints on its own path.
Collection schema
{
"type": "object",
"required": ["name"],
"properties": {
"name": { "type": "string" },
"note": { "type": "string" }
}
}Your site
await fetch('/spring-campaign/_data/entries') await fetch('/spring-campaign/_data/entries', { method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify({ data: { name, note } }) }) // Requests stay on the site origin.
Agent-first publishing
Works where you already use AI.
Agents upload the files they create and return the live URL. Choose the connection that fits your tool; use the dashboard when you want to handle a one-off yourself.
MCP
Web-based AI tools
Add the remote MCP server, sign in as yourself, and let the agent publish Markdown, HTML, or files directly.
CLI
Coding agents
Install the official skill. It uses the CLI to inspect local files, avoid unsafe inputs, and publish the right folder.
DROP
Manual one-offs
Drag in HTML, Markdown, a ZIP, multiple files, or a folder from the dashboard when no agent is involved.
Just published
acme.every.link/q3-dashboardQ3 Dashboard
14 files · published by Maya via Claude · deploy #42
FAQ
The usual questions.
What can I publish?
Anything static. A web AI tool can upload it through MCP; a coding agent can publish it with the skill and CLI; or you can manually drag in HTML, Markdown, a ZIP, files, or a folder. Markdown becomes a styled page.
Why not use my AI tool’s built-in publisher?
every.link gives your company more control after that work leaves the chat. Each site gets a stable company URL, its own access rules, a record of who published it, past versions, and one-click rollback.
The link stays with your company when people, models, accounts, or chats change. Teams can publish from an AI chat, a coding agent, the command line, or by dragging in files. Sites can also add uploads, data, and connections to other tools.
Does everyone share one agent token?
No. Each member authorizes their own AI connections, so publishes can be traced to both the person and the connected app. Admins can review and revoke access without seeing anyone’s secret.
Do my visitors need an account?
Only on sign-in-required sites. Public and password sites need no visitor account, and viewers are never billed.
Does republishing break links I have shared?
No. A site keeps its URL while each publish creates an immutable deploy you can inspect or roll back to.
What kind of database is it?
A JSON collection API for prototypes and internal tools. It is not positioned as a general-purpose production datastore.
Give your AI work an address.
Join the private beta and claim your account subdomain.