Skip to main content
This example demonstrates a powerful video editing workflow using VLM Run MCP tools. You can take a source video, trim it down to a specific segment, and then automatically generate and burn-in subtitles, all through simple conversational commands. This is perfect for creating short-form content, social media clips, and improving video accessibility.

​
Use Case Overview

Content creators and marketers often need to extract key moments from longer videos to share as highlights or promotional clips. This workflow automates that process. The agent first isolates a 15-second segment from a longer video based on user-provided timestamps. Then, it transcribes the audio from that clip and adds the text as synchronized, easy-to-read subtitles.

​
Workflow Steps

Try it out yourself with the following sequence of prompts (in Claude Desktop or Web):
  1. Load: “Load this video (https://storage.googleapis.com/vlm-data-public-prod/hub/examples/video/test_video.mp4)”
  2. Trim: “Ok, now trim this video from 00:30 to 00:45 s”
  3. Subtitle: “Ok, now add subtitles to this trimmed video.”
The agent uses a chain of video editing tools to fulfill this request: I’ll help you create a subtitled clip. First, I’ll load the video, then trim it to your specified time range, and finally, generate and add captions.
1

Load the source video

Use the put_file_url tool to load the source video into the system for processing.
{
  "url": "[https://storage.googleapis.com/vlm-data-public-prod/hub/examples/video/test_video.mp4](https://storage.googleapis.com/vlm-data-public-prod/hub/examples/video/test_video.mp4)"
}
Response:
{
  "id": "file_1821",
  "type": "video"
}
2

Trim the video to a segment

Use the trim_video tool to create a new video clip from the original source, specifying the start and end times.
{
  "video": {
    "id": "file_1821",
    "type": "video"
  },
  "start_time": "00:30",
  "end_time": "00:45"
}
Response: A new video object is created containing only the 15-second segment.
{
  "id": "file_9a81",
  "type": "video"
}
3

Add automatic subtitles

Use the add_video_captions tool on the trimmed video clip. The tool automatically transcribes the audio and burns the captions into the video frames.
{
  "video": {
    "id": "file_9a81",
    "type": "video"
  }
}
Response: A final, new video object is created that includes the subtitles.
{
  "id": "file_dde6",
  "type": "video"
}
4

Generate preview

Use the preview_object_ref tool to generate a shareable URL for the final, subtitled video clip.
{
  "obj": {
    "id": "file_dde6",
    "type": "video"
  }
}
Response:
[https://mcp.vlm.run/preview/file_dde6](https://mcp.vlm.run/preview/file_dde6)

​
Example Result

​
Original Clip (No Subtitles)

​
Final Clip (With Subtitles)

Video Trimming and Subtitling Results

​
Build Your Own Video Editing Pipeline

The tools for trimming, captioning, and manipulating videos can be combined to create a wide variety of automated editing workflows.
  • 📣 Marketing: Create social media-ready highlights from long-form interviews or product demos.
  • 🎓 Education: Generate short, subtitled educational clips from lengthy lectures or documentaries.
  • moderators Content Moderation: Isolate specific segments of a video for review based on transcript content.
  • 🎬 Media & Entertainment: Quickly assemble rough cuts or trailers by stitching multiple clips together.
  • ✍️ Accessibility: Batch process video libraries to add subtitles, making them accessible to a wider audience.

​
Try our MCP server today

Head over to our MCP server to start building your own document processing pipeline with VLM Run. Sign-up for access on our platform.
⌘I