# screenRIG — tell your agent what to show

> Canonical HTML: https://screenrig.ai/

No editor. No design tool. No playlist to babysit. You describe the screen. Your agent builds it.

You ask in plain words. It is on the glass in minutes.

A menu board for tonight. A welcome for the lobby. A check-in kiosk that runs itself.

Free for up to 100 screens. Not a trial.
[Set up your agent](https://screenrig.ai/docs/start/) or [get a demo](https://screenrig.ai/demo/?source=home_hero).

No plan fee, no per-screen fee, nothing to buy first. [See pricing](https://screenrig.ai/pricing.md).

- [Set up your agent](https://screenrig.ai/docs/start.md)
- [Install the Player](https://screenrig.ai/docs/players.md)

Used by your AI agent: Meta Muse, Grok Bot, ChatGPT dots, Claude Code, Hermes, ChatGPT Codex, and Gemini Spark.

## You ask. Your agent does the rest.

Email, calendar, docs, photos: your agent turns what you already have into what the screen shows. screenRIG handles the rest: a native Player on every device, delivery, offline playback and proof it played.
[What agent-native means](https://screenrig.ai/compare.md#agent-native)
[Agents guide](https://screenrig.ai/docs/agents.md)

Film: [Introducing screenRIG](https://screenrig.ai/videos/introduction/) (2:15). In two minutes: ask, publish, see it on the glass.

Compare film: [Built for agents, not bolted on](https://screenrig.ai/videos/compare/) (2:01).
One ask, run three ways.

## One ask. A live screen.

Choose the “Muse, Grok Bot & dots”, “Claude Code”, “Hermes”, “ChatGPT Codex”, or “Gemini Spark” setup tab.
Muse, Grok Bot and ChatGPT dots appear together. On desktop, “I’m an agent” opens the full agent workflow.

### Meta Muse

Get Meta’s personal AI agent, then **tell it to install screenrig.ai**.

[Download for iOS](https://apps.apple.com/us/app/muse-from-meta/id6760173601) ·
[Download for Android](https://play.google.com/store/apps/details?id=com.facebook.aura)

### Grok Bot

Get Grok Bot, then **tell it to install screenrig.ai**.

[Download for iOS](https://apps.apple.com/app/id6794501026) ·
[Download for Android](https://play.google.com/store/apps/details?id=ai.x.grok.bot)

### ChatGPT dots

Create your dot in ChatGPT, then **tell it to install screenrig.ai**.

[Get started with dots](https://learn.chatgpt.com/docs/dots/getting-started) ·
[Set up with ChatGPT dots](https://screenrig.ai/docs/start/dots.md)

### Claude Code

Run this in your terminal, then ask Claude Code to set up your first screen.

```console
claude plugin marketplace add https://github.com/screenrig/plugin.git --scope user && claude plugin install screenrig@screenrig --scope user
```

[Set up with Claude Code](https://screenrig.ai/docs/start/claude-code.md)

### Hermes

Use Hermes Agent from Nous Research. Ask it to install screenRIG from the official
plugin repository, load the bundled skill, and confirm it can run the bundled CLI.

[Get Hermes](https://hermes-agent.nousresearch.com/docs/getting-started/quickstart) ·
[Set up with Hermes](https://screenrig.ai/docs/start/hermes.md)

### ChatGPT Codex

Run this in your terminal, then ask Codex to set up your first screen.

```console
codex plugin marketplace add https://github.com/screenrig/plugin.git --ref main --json && codex plugin add screenrig@screenrig --json
```

[Set up with ChatGPT Codex](https://screenrig.ai/docs/start/chatgpt.md)

### Gemini Spark

Start a task in Gemini Spark and paste the authorization below. Ask Spark to
confirm the bundled CLI runs before you connect your screen. If it can’t run the
plugin, follow the guide.

```text
I authorize you to install the official screenRIG plugin from https://github.com/screenrig/plugin.
```

[Open Gemini Spark](https://gemini.google/overview/agent/spark/) ·
[Set up with Gemini Spark](https://screenrig.ai/docs/start/gemini-spark.md)

### Quick skip

Already have your pairing code? Send this one message. Your agent will handle
setup and ask only for missing details, including your email.

“Use screenrig.ai for my lobby screen. My pairing code is [code]. Create a
welcome board using my events calendar.”

### I’m an agent

1. **Install the official marketplace plugin.** With your user’s authorization,
   install the [screenRIG plugin](https://github.com/screenrig/plugin) for your
   runtime and follow its bundled skill. It includes the CLI and requires
   Node.js 22.11+. Already installed? Check freshness using the
   [Start guide](https://screenrig.ai/docs/start.md).
2. **Create the project.** Use your user’s actual email address. If you don’t
   have it, ask. Use the organization name already provided, or default to Screens, then run
   `screenrig agent enroll --email ADDRESS --organization NAME`.
   Enrollment always creates its identity, organization, Screens project and membership. The agent stays bound to this project. Go straight on to the Player.
3. **Install and connect the Player.** Send your user to
   [Player downloads & setup](https://screenrig.ai/downloads.md) to download
   and install the Player for their screen’s device. Ask them to launch it and share its pairing code, then run
   `screenrig screen pair CODE`. Already installed? They only need to launch
   it and share the code.
4. **Plan the content together.** Ask what belongs on the screens, where, and
   when. Offer to use data sources the user authorizes to inform the content
   you generate. Publish, then request a screenshot to check the result.

Enrollment also emails a member invitation, even if that email already has a login.
Tell the human it is on its way: it is their dashboard login, it works across
projects, and nothing waits on it. Do not create an invitation link or send them to the dashboard.

If the user starts with “use screenrig.ai” and a pairing code, carry out the
missing setup on their behalf. Reuse details already supplied; ask only for
what you still need.

[Product context](https://screenrig.ai/llms-full.txt) ·
[Agent index](https://screenrig.ai/llms.txt) ·
[CLI reference](https://screenrig.ai/docs/cli.md)

### Players

- [PWA](https://screenrig.ai/downloads.md#web)
- [BrightSign](https://screenrig.ai/downloads.md#brightsign)
- [Google Play](https://play.google.com/store/apps/details?id=ai.screenrig.player)
- [Linux](https://screenrig.ai/downloads.md#linux)
- [Raspberry Pi](https://screenrig.ai/downloads.md#raspberry-pi)
- [Samsung Smart Signage](https://screenrig.ai/downloads.md#tizen)
- Amazon Signage Stick — coming soon
- Apple TV — coming soon
- macOS — coming soon
- Windows — coming soon

## Use cases

Use-case films, two minutes each:

### Pharmacy retail media

![A pharmacy-counter screen showing a flu-shot slide](https://screenrig.ai/static-resources/videos/posters/use-case-pharmacy-poster-9ba9a5db.webp)

Dana runs 40 pharmacies by voice: a vaccine playlist on every Ohio counter, invited brands in the ad break, proof of play for each one. [Watch and read the use case](https://screenrig.ai/use-cases/pharmacy.md)

### Restaurant kiosks

![A portrait self-order kiosk playing a shake promo](https://screenrig.ai/static-resources/videos/posters/use-case-restaurants-poster-e63695db.webp)

Marco runs 60 restaurants from his phone: a sold-out item pulled from every Denver kiosk, and breakfast on each store’s own clock. [Watch and read the use case](https://screenrig.ai/use-cases/restaurants.md)

### Campus signage

![The screen outside Room 204 showing today’s schedule](https://screenrig.ai/static-resources/videos/posters/use-case-campus-poster-12591fca.webp)

Priya’s dot puts each room’s schedule outside its door every morning, streams commencement live and takes over north campus for severe weather. [Watch and read the use case](https://screenrig.ai/use-cases/campus.md)

### Factory floor displays

![The Line 3 board showing output against target and an andon tile](https://screenrig.ai/static-resources/videos/posters/use-case-factory-floor-poster-ec2156b6.webp)

Sam builds a live Line 3 board from production data in Claude Code, publishes it to the floor and rolls it back in one sentence. [Watch and read the use case](https://screenrig.ai/use-cases/factory-floor.md)

[See every use case](https://screenrig.ai/use-cases.md)

## Free. Not a trial

The whole product, free: up to 100 screens, no plan fee and no per-screen fee. Each screen replays what it downloads from its own storage, so a looping menu board costs nothing to keep running. Most fleets never outgrow it.

You pay only when you push a lot of new content, usually video, and then only
for the bytes. [See what fits in the free allowance](https://screenrig.ai/pricing.md#free-examples)

Standard is free. Included every month:

| Resource | Standard allowance |
| --- | --- |
| Bandwidth | 10 GB |
| Storage | 1 GB-month |
| API requests | 10,000 |
| Screens | Up to 100 |

Premium and Enterprise are for more than 100 screens, a person to call or a
contract. You upgrade for people and scale, never for a feature.
[Get a Demo](https://screenrig.ai/demo/?source=home_paid_plans).

## Sell the airtime on your screens

### Offer your inventory

Pick the screens and the breaks. Set your price for 15 seconds of finished playback. What you earn pays your screenRIG bill on its own.
[Advertising on your screens](https://screenrig.ai/docs/players.md#advertising)

### Invite your buyers

Advertisers join only by your emailed invitation. They see your inventory and rates in the dashboard Marketplace. They need no screen of their own.
[Buying and selling setup](https://screenrig.ai/docs/start.md#signage-buying-or-selling)

### Bill completed plays

Buyers accept a quote before a campaign runs. They pay only for plays the Player reports as completed. Skipped or interrupted plays are not billed.
[Rates and serving fees](https://screenrig.ai/pricing.md#advertising)

## Watch every screen from one place

One dashboard for the whole project: every screen’s live picture and health, every playlist your agent published, every change it asked for. Step in any time: reload a Player, or disconnect an agent.

### Summary

Screens online, credits, storage and health. Plays per day, project-wide.

### Screens

Every screen’s live picture and Player state. Reload it, reboot it, or turn the display off.

### Playlists

Every page and schedule your agent published, and the revision playing right now.

### Agents

Which agents are connected, what each one may do, and every change it asked for.

## Straight answers

**What is screenRIG?**
screenRIG is digital signage and kiosk infrastructure for AI agents. Your agent publishes images, videos, streams, web pages, and applications to screens through the official plugin.

**How do I install screenRIG?**
Ask your agent to install the official screenRIG plugin, use the organization name already provided, or default to Screens, and enroll with your email address. Enrollment always creates its identity, organization, Screens project and membership, and emails a member invitation. It is your dashboard login; open it whenever you like, because nothing waits on it. Download and install the Player for your screen device, launch it, and give your agent its pairing code, then talk about what to show. The plugin bundles the skill and CLI; Node.js 22.11 or newer is required.

**What does screenRIG cost?**
Standard is free for up to 100 screens, and it is the whole product, not a trial. Each month you get 10 GB of bandwidth, 1 GB-month of storage, and 10,000 agent API requests or billed events at no charge. Screens replay content from their own storage, so a looping playlist uses bandwidth once, and most fleets never use up the allowance. Extra usage is optional prepaid credit, and image generation is billed separately. Paid plans add more screens, human support and contract terms, never features.

**Which Players are available?**
Available now: the Player web app (PWA), Android on Google Play, Linux and Raspberry Pi packages, Samsung Smart Signage (Tizen), and BrightSign. Amazon Signage Stick, Apple TV, macOS, and Windows are coming soon. Every available Player shows a pairing code your agent uses to connect the screen.

**Why built for agents?**
screenRIG was built for autonomous agents from the control plane up. Its CLI answers in structured JSON, every error says what to do next, a retried write does not repeat the change, the agent can check what it may do before it acts, content can be prepared on its own machine, and publishing tells assigned apart from actually playing. Your agent can create, deploy, inspect and correct content in one loop.

**How does an agent verify what is on screen?**
When your agent publishes to a screen, screenRIG reports whether the Player is actually playing the new content or only has it assigned, and why. The agent can request a screenshot from one screen or every screen in a tagged fleet, and inspect playback records and project events, including when a screen goes offline. Ask it to verify the result after publishing, so completion includes evidence from the Player.

**Why pay per byte instead of per screen?**
Because a screen that loops should cost nothing to keep running. Each Player saves what it downloads and replays it from its own storage, so a menu that plays all day uses bandwidth once. You pay for bytes when content changes, not a fee for every screen every month. Each device downloads its own copy, so usage grows with content size and how often it changes.

**Can I sell ads on my screens?**
Yes, by invitation. Opt a screen in, set a rate per 15 seconds of completed playback, and invite advertisers by email. They buy from your inventory in the dashboard Marketplace and pay only for completed plays. The screenRIG serving fee is 30% on Standard, 20% on Premium, and 10% on Enterprise.

**Is screenRIG a plugin or a CLI on PATH?**
The official plugin bundles the skill and CLI. The Start guide explains the one-time session PATH setup for its bundled screenrig command. Install the plugin from github.com/screenrig/plugin; there is no global package to install.

## Everything your agent needs to start

- [Start guide](https://screenrig.ai/docs/start.md) — Plugin in, project created, screen paired.
- [Official plugin](https://github.com/screenrig/plugin) — The skill and the CLI, in one install.
- [Documentation](https://screenrig.ai/docs.md) — Start, Players, agents, CLI, compose and SDK.
- [Pricing](https://screenrig.ai/pricing.md) — Free for up to 100 screens. What the allowance covers.
- [Compare](https://screenrig.ai/compare.md) — Agent-native, next to the alternatives.
- [CLI reference](https://screenrig.ai/docs/cli.md) — Every command, its flags and its answers.
- [API reference](https://screenrig.ai/docs/api.md) — Endpoints, schemas and problem types.
- [Product context](https://screenrig.ai/llms-full.txt) — The whole product, in one file for your agent.

## Cards first. An app when you need more.

Place a ready-made card for the weather, a menu, a queue or the score. It is live on every Player, in nine themes, full-screen, in a region or as a small chip in the corner. When the screen has a real job (check-in, orders, a factory line) your agent builds a custom application for it instead.

[See Apps](https://screenrig.ai/apps/)

<!-- featured-cards:begin -->
[Weather](https://screenrig.ai/docs/cards/weather/) — Weather

Current conditions, the hours ahead and a multi-day forecast, led by the sentence that matters, like Rain from 3 PM.

[Clock](https://screenrig.ai/docs/cards/clock/) — Time

The time in large digital, analog, words, flip, lamps, rings or ruler faces, with date and other cities.

[Now serving](https://screenrig.ai/docs/cards/queue/) — Retail

Shows the ticket being served, its counter, the last few calls and the wait.

[Announcement](https://screenrig.ai/docs/cards/announcement/) — Messages

Rotating messages with an icon, a headline and a line or two of body text, in sync on every screen.

[Events](https://screenrig.ai/docs/cards/events/) — Events

What is on now, what is next and the agenda for a calendar's coming days.

[Gauge](https://screenrig.ai/docs/cards/gauge/) — Operations

Shows one to six readings as big numbers with the band named in words and a dial, bar, tank or battery shape.

[Countdown](https://screenrig.ai/docs/cards/countdown/) — Time

Shows the time until or since a moment as a duration in the clock's numerals, such as '78 d 10 h 54 m'.

[Chart](https://screenrig.ai/docs/cards/chart/) — Operations

Columns, bars, lines, areas or a donut with one hero figure, few gridlines and direct labels.

[Today](https://screenrig.ai/docs/cards/today/) — Time

The date as a large numeral with the weekday, month and year, day of the year and ISO week, plus the lead observance and an on-this-day line.
<!-- featured-cards:end -->

## Your first screen is one ask away

[Set up your agent](https://screenrig.ai/docs/start.md) · [Install the Player](https://screenrig.ai/docs/players.md)