Motion graphics videos from your AI agent
Connect Claude, ChatGPT, Claude Code, Codex, Cursor or VS Code to iArt’s MCP server. Your agent describes the video; iArt writes it, adds voice-over and renders it. Back come an MP4 and a share link. A REST API does the same.
Beta. Pro, Ultra and Max plans.
Claude app
In claude.ai or the Claude desktop app: Customize → Connectors → Add → Custom connector. Paste this URL, choose Register automatically for the OAuth client, then Connect and sign in to iArt. No API key needed.
Replace YOUR_API_KEY with a key from the API page in the app.
https://app.iart.ai/api/mcpChatGPT
In ChatGPT: Plugins → Add → Add custom MCP server. Name it iArt, paste this URL, set Authentication to OAuth, then sign in to iArt. No API key needed.
Replace YOUR_API_KEY with a key from the API page in the app.
https://app.iart.ai/api/mcpClaude Code
Run in a terminal.
Replace YOUR_API_KEY with a key from the API page in the app.
claude mcp add --transport http iart https://app.iart.ai/api/mcp --header "Authorization: Bearer YOUR_API_KEY"Codex
Run in a terminal. Codex reads the key from IART_API_KEY, so keep it set where you start Codex (e.g. in your shell profile). Or skip the key: in the Codex app, add an MCP server of type Streamable HTTP with the URL https://app.iart.ai/api/mcp and sign in to iArt when asked.
Replace YOUR_API_KEY with a key from the API page in the app.
export IART_API_KEY=YOUR_API_KEY
codex mcp add iart --url https://app.iart.ai/api/mcp --bearer-token-env-var IART_API_KEYCursor
Add to .cursor/mcp.json in your project (or ~/.cursor/mcp.json).
Replace YOUR_API_KEY with a key from the API page in the app.
{
"mcpServers": {
"iart": {
"url": "https://app.iart.ai/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}VS Code
Add to .vscode/mcp.json in your workspace.
Replace YOUR_API_KEY with a key from the API page in the app.
{
"servers": {
"iart": {
"type": "http",
"url": "https://app.iart.ai/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Any other MCP client: Streamable HTTP at https://app.iart.ai/api/mcp, with the key in an Authorization: Bearer or x-api-key header. Then ask your agent for a video, for example “make a 15-second launch video for my app with iArt”.
Seven tools
Full API referencecreate_video
POST /api/v1/videos
Starts a motion graphics video from a written brief: prompt (required), aspect_ratio (16:9, 9:16 or 1:1), voiceover (on by default), model, resolution (720p by default; 1080p on Pro, up to 4K on Ultra and Max) and files (up to 10: public links, up to 10 MB for images and 100 MB for video, or files from upload_file). Returns the video id.
get_video
GET /api/v1/videos/{id}
Status: writing, rendering, done, failed or stopped. When done: an MP4 link (works for 24 hours, so download it), an unlisted share link that does not expire, and a link that opens the project in the iArt editor.
revise_video
POST /api/v1/videos/{id}/revise
Changes a finished video from a description of what to change (and more files, if you like), or retries a failed or stopped one. It is made again and charged again.
stop_video
POST /api/v1/videos/{id}/stop
Stops a video being made, for example when the brief was wrong. What it used so far is charged; it can be retried later.
list_videos
GET /api/v1/videos
The videos you made with the API, newest first (10 by default, up to 50), each with the opening of its brief and when it was made, so your agent can find an earlier one.
get_account
GET /api/v1/account
Your plan, the credits available and left for new videos, how many videos you can make at a time and how many are in progress, and the highest resolution.
upload_file
POST /api/v1/files?name=…
Adds a file from your computer to use in a video: a logo, screenshots, product photos, a voice recording, a PDF (up to 4 MB; for larger files, put a public link in files instead). Over MCP your agent gets an upload link and sends the file with curl, no key needed; over REST you send the file as the body. Either way you get a file_url for files. Every file is checked before it is used: it must be the type its name says, and an SVG may not contain scripts or links to other files.
REST
# Upload a file to use (optional; answers with its file_url)
curl -X POST "https://app.iart.ai/api/v1/files?name=logo.png" -H "Authorization: Bearer YOUR_API_KEY" --data-binary @logo.png
# Start a video (answers at once with its id)
curl -X POST https://app.iart.ai/api/v1/videos \
-H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" \
-d '{"prompt": "A 15-second launch video for my app, with logo.png: …", "files": ["FILE_URL"]}'
# Check it every 30-60 seconds until status is done, failed or stopped
curl https://app.iart.ai/api/v1/videos/VIDEO_ID -H "Authorization: Bearer YOUR_API_KEY"
# Change a finished video (or retry a failed or stopped one, without instructions)
curl -X POST https://app.iart.ai/api/v1/videos/VIDEO_ID/revise \
-H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" \
-d '{"instructions": "Make the logo bigger"}'
# Stop a video being made
curl -X POST https://app.iart.ai/api/v1/videos/VIDEO_ID/stop -H "Authorization: Bearer YOUR_API_KEY"Limits and pricing
Videos use your credits at the same prices as the app. A video is refused up front when your credits can’t cover it on top of the ones in progress.
- Pro
- 3 videos at once · up to 1080p
- Ultra
- 6 videos at once · up to 4K
- Max
- 15 videos at once · up to 4K
Supported models and pricing
- model
- Input
- Output
- deepseek-4.1
- $0.45 /1M tokens
- $1.8 /1M tokens
- haiku-5.5
- $0.15 /1M tokens
- $0.75 /1M tokens
- sonnet-5.5 (default)
- $3 /1M tokens
- $15 /1M tokens
- opus-5.5
- $6 /1M tokens
- $30 /1M tokens
- fable-5.2
- $12 /1M tokens
- $60 /1M tokens
- gemini-4-argon
- $3 /1M tokens
- $15 /1M tokens
Including the service fee. Rendering: $0.09 per scene and $0.045 to combine the scenes, at any resolution. Per key: 300 status checks and 60 creates or revisions a minute.
Questions
What is the iArt MCP server?
A remote MCP server that lets an AI agent such as Claude, ChatGPT, Claude Code, Codex, Cursor or VS Code make motion graphics videos with iArt: kinetic typography, explainers, logo reveals, data animations and promos, with AI voice-over. Your agent describes the video, iArt writes and renders it, and the agent gets back an MP4 and a share link. The same tools are available as a REST API.
Which plans include the API?
Pro, Ultra and Max. Create a key in the iArt app under API (in your account menu). Videos use your account’s credits, exactly as videos made in the app do.
How much does a video cost?
Videos use your credits at the same prices as the app: the model’s tokens (from $0.15 in and $0.75 out per 1M tokens with haiku-5.5; sonnet-5.5, the default, is $3 and $15), plus $0.09 per rendered scene and $0.045 to combine them, at any resolution. What a video costs depends on its length and how much the agent writes. A video is refused up front when your credits can’t cover it on top of the videos you have in progress.
How long does a video take, and what do I get?
A few minutes. You get an MP4 (720p by default; 1080p on Pro, up to 4K on Ultra and Max), an unlisted share link anyone with the link can watch, and a link that opens the project in the iArt editor, where you can keep working on it.
How many videos can I make at once?
Pro 3, Ultra 6 and Max 15 at a time, counting videos you make with Batch in the app too. The API is in beta. Each key also has a rate limit: 300 status checks and 60 creates or revisions a minute.
Is there a free way to do this?
The open-source motion graphics skills teach your agent to write and render animations itself, for free. The API is for when you would rather have iArt run the whole pipeline: writing, voice-over, music and rendering. Browse the skills.