compose_projectLABSCO SUMMARY
One call for a whole media brief, rather than a model picker and a prompt.
The interesting part is what you do not have to decide: compose_project takes the description, including variations, and Glif resolves which models run in what order. Because it answers with a job id rather than the media, anything built on it has to poll — the finished files arrive as links to a CDN, not as bytes in the conversation, which is what keeps a batch of variations affordable to ask for.
What it is
A hosted media-generation server at https://glif.app/api/mcp: generate images, video and audio, transcribe, render HTML, search the web, run code, and chain multi-step media operations from any MCP client.
What you get
- compose_project takes a plain-language brief — including a series of variations — and Glif chooses the models and chains the steps itself
- The job returns a job_id straight away; get_job_status returns the finished media as resource_link blocks pointing at CDN URLs
- Read and upload helpers around those two, with tools/list as the authoritative set for whatever your client sees
- Streamable HTTP with OAuth 2.1 and dynamic client registration — sign in with your Glif account, no API key to place
- One-click install snippets for Claude, Claude Code, Cursor, VS Code, ChatGPT and Codex
Requirements
A Glif account and a client that can reach a remote MCP server: paste https://glif.app/api/mcp as a custom connector and approve the browser sign-in on first connect. Generation spends credits from that account; the read-only tools do not. The locally run npm package is deprecated, so a config that launches it over stdio should move to the URL. MIT covers this repository, which holds the registry metadata; the hosted server itself is not open source.
Setup effort
Add the hosted endpoint, then sign in — paste it into your client, then approve the OAuth sign-in
USE CASES
01
Getting a set of variations rather than one image
Describe the series you want in one brief and let the server chain the steps, then collect the finished files from the job once it reports done.
02
Adding media generation to an agent without wiring up model APIs
One remote connector covers image, video and audio generation, so there is no per-model key to manage or SDK to install.
03
Waiting on a long render without blocking the session
The call returns a job id immediately; poll for status and pick up the CDN links when the work finishes.
2 tools
Listed in project docs. Your client may show a different list.
get_job_statusBEFORE YOU INSTALL
What to know first
Tools you get
2 tools named in the project's documentation. Only compose_project and get_job_status are named; call tools/list or see https://glif.app/mcp for the full set, which also includes read and upload helpers.
INSTALLATION
Connect your client
Server setup
ENDPOINT URL
https://glif.app/api/mcpAdd this hosted endpoint to your MCP client, then sign in.
MCP Clients
Claude Code
CLAUDE CODE · TERMINAL
claude mcp add --scope user --transport http glif "https://glif.app/api/mcp"Cursor
CURSOR · JSON
{
"mcpServers": {
"glif": {
"url": "https://glif.app/api/mcp"
}
}
}VS Code
VS CODE · TERMINAL
code --add-mcp '{"name":"glif","type":"http","url":"https://glif.app/api/mcp"}'Codex
CODEX · TERMINAL
codex mcp add glif --url "https://glif.app/api/mcp" ; codex mcp login glifAny MCP client
ANY MCP CLIENT · JSON
{
"mcpServers": {
"glif": {
"url": "https://glif.app/api/mcp",
"type": "http"
}
}
}Claude.ai
Settings → Connectors → Add custom connector.
Paste https://glif.app/api/mcp and approve the Glif sign-in that opens in your browser.
ChatGPT
Enable developer mode, then Settings → Apps & Connectors → Add new connector.
Paste https://glif.app/api/mcp and pick OAuth.
