MCP Connector
LiveRenderly is now available as an MCP (Model Context Protocol) connector — a remote server that brings your entire Renderly library into your AI assistant. Connect it to Claude on the web, desktop, or Claude Code, then create videos by just talking: "Render my real-estate template for these 10 listings." Claude discovers your templates and projects, fills in the variables, fires the renders, and hands back the video links. It's the fastest way to generate personalized video at scale — your AI agent does the work, drawing from your own Renderly account and credits.
Triggers & Actions
Events that start your workflow and actions you can automate with MCP Connector.
Browse Your Library
Ask your AI assistant what's available and it lists your Renderly projects and the public system templates — no clicking through the dashboard. "What Renderly templates can I use?"
Discover Variables
Claude inspects any template or project and surfaces exactly what you can personalize — headlines, prices, addresses, background clips — along with their current values.
Create a Render
Render from a template or project with your own replacement values, all in one sentence. Perfect for batch and personalized output: "Render my Sale template for each of these 8 product names." Uses your Renderly credits.
Track Output
Claude polls the render job and hands back the status, progress, and the final video URL when it's ready. "Is that render done? Give me the link."
How It Works
Get up and running with MCP Connector in a few simple steps.
Make sure you have a Renderly account
The connector acts as your Renderly account, drawing from your credits and your own templates and projects. If you don't have one yet, sign up at renderly.video first.
Add the connector to Claude
In Claude.ai, add a custom connector pointing at https://renderly.video/api/mcp. On Pro/Max, add it directly under Connectors. On Team/Enterprise, an Owner adds it once under Organization settings → Connectors and members enable it.
Or add it in Claude Code
From the terminal, run: claude mcp add --transport http --scope user renderly https://renderly.video/api/mcp — then run /mcp to sign in. Prefer a key? Append --header "Authorization: Bearer rnd_YOUR_API_KEY".
Sign in with Renderly
Authorize the connector with OAuth ("Sign in with Renderly") or a Renderly API key. Use the email that owns your Renderly account so the tools can find your library.
Ask what you can do
Confirm the connection by asking Claude to list your templates or projects. It will show what's available and which variables each one exposes.
Render in plain language
Describe the video you want — "Render my Listing Promo with address '123 Ocean Dr' and price '$1.2M'" — and Claude fills the variables, starts the render, and returns the finished video link.
Example Workflow
Personalized real-estate videos at scale — just by asking
A realtor has 10 new listings and a Renderly "Listing Promo" template. Instead of editing 10 videos by hand, they hand the whole batch to Claude in a single message.
- 1The realtor pastes 10 listings (address, price, photo) into Claude and asks it to render the Listing Promo for each
- 2Claude uses the connector to read the template's variables — address, price, and background image
- 3For every listing, Claude calls Create Render with the right replacement values, kicking off 10 renders
- 4Renderly processes each render and Claude tracks the job status for all of them
- 5Claude hands back 10 finished video URLs, ready to post to listings, social, or email — no dashboard, no manual editing
Frequently Asked Questions
Quick answers about setup, billing, and troubleshooting.
