Skip to main content

Documentation Index

Fetch the complete documentation index at: https://rendobar.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

In this tutorial we’ll use Claude Desktop to caption a local video. The agent uploads it to Rendobar, runs an FFmpeg job that burns subtitles into the video, and returns a download URL. End-to-end takes about eight minutes. You will need:
  • A Rendobar API key (see Installation step 1)
  • Claude Desktop with @rendobar/mcp configured (see Installation step 2)
  • A short .mp4 file on your laptop
  • An .srt subtitle file at the same location
If you don’t have an .srt handy, drop this into subs.srt next to your video: 
1
00:00:00,000 --> 00:00:05,000
This is a test caption from Rendobar.

2
00:00:05,000 --> 00:00:10,000
The agent burned it into the video.

​
Step 1: Confirm the agent sees Rendobar

Open Claude Desktop and ask:
What MCP tools do you have for Rendobar?
The agent lists upload_file, submit_job, get_job, list_jobs, cancel_job, get_account. If it doesn’t, the MCP isn’t loaded — re-check your config and restart Claude Desktop.

​
Step 2: Check the balance

Ask:
Check my Rendobar balance.
The agent calls get_account and reports something like *“29.42,freeplan,100MBmaxupload"∗.We′regoingtospendaround29.42, free plan, 100 MB max upload"*. We're going to spend around 0.05.

​
Step 3: Hand it the files

Tell the agent the goal:
Burn the captions in ~/Videos/clip.srt into ~/Videos/clip.mp4 and give me the result. Use Rendobar.
The agent runs three tool calls in sequence:
  1. upload_file({ path: "~/Videos/clip.mp4" }) returns a downloadUrl
  2. upload_file({ path: "~/Videos/clip.srt" }) returns a second downloadUrl
  3. submit_job with type: "raw.ffmpeg", the two URLs as inputs, and an FFmpeg command that uses -vf subtitles
You see each call as it happens in Claude Desktop’s tool-call panel.

​
Step 4: Wait

The agent polls get_job every few seconds. A 30-second clip with subtitles takes about 20 to 40 seconds to render.

​
Step 5: Download

When get_job returns status: "complete", it includes an outputUrl. The agent shows you a clickable link. Open it in your browser and the captioned video downloads.

​
What just happened

You handed the agent two local files and a goal. It uploaded each to ephemeral storage, submitted a job to Rendobar’s FFmpeg infrastructure, and polled until done. The whole flow used six tool calls and one English sentence from you. The output file lives at cdn.rendobar.com for 24 hours by default, then auto-deletes. The uploaded source files auto-delete in the same window.

​
Next