Where humans and your agents
work on dashboards together.

Cursor, Codex or any other agent publishes the dashboard to Display. Its URL, versions and comments stay when you change agents.

Inside your agent

The demo shows an agent publishing a doc to Display from a chat, then a teammate opening its URL, selecting Edit, changing a number on the page and saving v2, a colleague commenting on v2, the agent publishing v3 at the same URL and resolving the comment, and the doc being shared with someone outside the company.

Used by top teams at

Customer.ioHighLevel6senseKogan.comSohonetHeliosXSmartwayFairingPluralOutdoor CapTagadaModos

The agentic publishing problem

An agent publishes a financial analysis for the whole company when only three reviewers should see it. Another finishes a dashboard, but the next agent cannot find the source or the feedback. Those are publishing failures, not model failures. The doc needs one URL with explicit access, retained versions, source and comments. In Linear's March 2026 announcement, Karri Saarinen reported that a growing share of new issues are now agent-authored, and that the volume of agent-completed work is climbing fast. Agents write reports, produce analyses, generate dashboards and output HTML documentation at a scale no human team could match. All of it needs to reach the right people at a stable URL, without a publishing step from the engineer who ran the agent.

  1. 01

    Email and Slack

    No stable URL, no interactivity, re-sent every time.

  2. 02

    GitHub Pages

    Needs a git push, and excludes non-engineers without Enterprise Cloud.

  3. 03

    Confluence and Notion

    Can't render the HTML agents produce.

  4. 04

    Vercel and Netlify

    Need a deployment project. Company sign-in needs a higher plan.

Display was designed from the start to be callable. The publish action is a function, not a workflow.

Three ways an agent publishes

  1. 01

    MCP

    For Claude, Claude Code and any MCP-compatible agent. The agent calls the Display publish tool, and the doc's URL comes back in the same call.

  2. 02

    REST

    Any agent, any language. Send the content and a name to the Display REST API with your API key. The response carries the URL.

  3. 03

    A request

    Ask the agent to publish the report to Display. For agents that write files as they work. The agent publishes whatever the run wrote to disk.

Who can open it

  1. People and groups you choose

    Every new doc starts private. Share it with people and groups, and invite clients or partners by email with a role: Viewer, Commenter, Contributor or Editor.

  2. Everyone at your company

    Open a doc to your whole organization. On Solo and higher plans, colleagues sign in with the Google or Microsoft account they already use.

  3. Anyone with the link

    Make a doc public when anyone should open it, with no sign-in.

  4. Know who opened it

    On Team and Business, see who opened each doc, and when.

The right pricing model for agent-scale publishing

Per-seat pricing breaks at agent scale. If your agent publishes 50 reports a week to an audience of 200 colleagues, per-seat pricing charges you for each viewer regardless of how many reports they actually read. The 201st colleague triggers a pricing jump. Display charges per organization. You never pay for seats, and viewers are unlimited on every plan.

Per-seat pricing (Cloudflare Access)DisplayDisplay
100 viewers, 10 docs/week~$350/mo$59 a month, billed yearly
200 viewers, 50 docs/week~$1,050/mo$59 a month, billed yearly
500 viewers, 200 docs/week~$3,150/mo$59 a month, billed yearly
Scale to zero (no views this week)Still charging per seat$59 a month, billed yearly
Get your API key free

Cloudflare Access is $7 per user per month, with the first 50 users free – so 100 viewers is 50 billable seats.

What agents publish

Recurring analysis agents
– Weekly competitive reports, market monitoring dashboards, financial summaries. The agent runs on a schedule and publishes the next version at the same URL. Stakeholders bookmark the stable URL.
Scheduled-build agents
– Nightly research summaries, weekly competitive landscape regenerations, scheduled market-map updates. The agent publishes the next version after every run, at the same URL. The Slack notification links to an authenticated URL.
Research agents
– Deep research outputs, synthesis reports, literature reviews. Produced programmatically and published to a team-accessible URL without a human copying the content.
Tool-building agents
– Cursor, Codex, and Claude Code agents that scaffold internal tools. The agent builds, then publishes the output. Non-engineers get a usable URL.

Add Display to your agent stack

Claude and other MCP hosts
Add Display from the Claude connector directory, or paste https://api.display.dev/v1/mcp into the host's "Add custom connector" field. Sign in with your work email. There is no key to manage.
What the agent can do
Publish a doc and update it at the same URL. Search and read every doc you can open. Share docs, and read and resolve comments. The MCP page lists every tool.
Custom agents
The REST API has the same actions. Create an API key in the dashboard.

Never pay for seats

No per-API-call charges. No per-publish charges. No per-viewer charges. Flat.

Solo

$9/mo · $108 billed yearly

Includes

  • 1 Editor or Contributor
  • 5 GB storage
  • Unlimited gated documents
  • Unlimited Viewers and Commenters
  • + 8 more
TeamMost popular

$59/mo · $708 billed yearly

Everything in Solo, plus

  • 150 Editors and Contributors
  • 100 GB storage
  • Shared Spaces
  • Viewer identities
  • + 1 more
Business

$199/mo · $2,388 billed yearly

Everything in Team, plus

  • 1 TB storage
  • Audit logs, 365 days
  • Additional email domains
  • Public and external sharing controls
  • + 1 more
Free

$0

Includes 1 Editor or Contributor · 3 gated documents · 50 MB storage · Random subdomain

Get started
Enterprise

Custom

Need more than 150 Editors and Contributors, or a custom contract?

Book a call
Built and backed by people behindPipedrivePactum AIWise

Features

  • Authentication

    Viewers click a link and sign in with their Google or Microsoft account, or a one-time password. No app to install. No account to create. No IT ticket.

  • Comments and edits

    Teammates comment on any line of the page. Your agent reads the comments, publishes the next version at the same URL and resolves the threads it answered. People with edit access select Edit and change the doc right on the page.

  • Publish in 15 seconds

    Ask your agent to publish the file to Display. It returns a URL. No git repo, no deploy pipeline, no project to configure.

  • URLs that never change

    Every doc gets a URL that keeps working. Share in Slack, link in Notion, paste in email. It still works six months later.

  • Unlimited viewers

    No per-seat pricing at any tier. Share with your PM, exec, legal team, or designer for the same flat price.

  • Any agent

    Cursor, Codex, Claude Code or any other agent that writes HTML or Markdown. Each person can use their own agent and AI plan, personal or company, on the same doc.

A full workspace behind every URL

Share with your team and clients

Invite people by email and give each one a role.

Go back to any version

Each publish and each edit saves a version. The link stays the same.

Search for people and agents

You and your agents search every doc you can open.

See who opened it

Each doc shows who opened it, and when.

Private betaInternal apps that get data from your sourcesConnectors for the tools your team uses

Get started free

Get an API key and publish from an agent in minutes, no card · Keep the URL when you upgrade · Unlimited viewers on every plan

Common questions

Who can open a doc?

A new doc is private by default. You can share it with people and groups, open it to everyone in your company, or make it public. People sign in with Google, Microsoft or a one-time password before a private or company doc opens. Enterprise adds SAML sign-in, for example with Okta.

How do people and agents work on a doc together?

Teammates comment on any line of the page. People with edit access select Edit and change the doc right on the page. Your agent reads the doc and its comments, publishes the next version at the same URL, and resolves the threads it answered. Each save or publish adds a new version. On Solo and higher plans, you can open and restore earlier versions.

Can people outside our company contribute?

Yes. Invite them by email to a doc or a folder, and give them a role: Viewer, Commenter, Contributor or Editor. They sign in with a one-time password, Google or Microsoft. They do not join your organization.

Can people on personal Claude plans share docs with our Claude Team or Enterprise users?

Yes. Display does not check anyone's AI plan. Everyone signs in to your Display organization with their work email, whether their agent runs on Claude Pro, Max, Team or Enterprise. They all publish to and open the same docs.

Which agents work with Display?

Any agent that writes HTML or Markdown, for example Claude Code, Claude, ChatGPT, Codex and Cursor. Agents connect through the Display MCP server, the CLI or the API. Each person can use a different agent on the same doc.

What can I keep in Display?

HTML and Markdown files open as pages: reports, dashboards, specs, prototypes and slides. Folders also store other files, for example images, PDFs and data files, so people and agents find everything in one place.

Can Display replace Notion or Confluence?

Yes, for the docs that your team and your agents write: specs, plans, reports and wikis. Spaces and folders organize them, search finds them, and people comment and edit right on the page. To move existing pages, export them from Notion or Confluence as HTML or Markdown, and ask your agent to publish them.

Can I see who opened my docs?

Yes. On Solo and higher plans, every doc shows its view count. Team and Business also show who opened each doc, and when.

Can a published page call other services?

Yes. An Owner adds the domains of those services to the domain allowlist in Settings. Your published HTML pages can then get data from those domains and send data to them, for example from a form to a webhook. The other service must also accept the request. Display does not add a login or an API key, so keep secrets out of the page. The Business plan includes the domain allowlist.

Does Display use my content to train AI models?

No. We do not use customer content to train or fine-tune any AI or machine-learning model.