← Back to VidLet

Documentation

Auto-captions and 16 video tools that run on your machine. Drive them from the terminal, or hand them to an AI agent.

Install

Requires WSL + Node.js 18+ + FFmpeg. whisper.cpp auto-downloads on first caption use.

sudo apt install ffmpeg
npm i -g @spark-apps/vidlet
vidlet caption video.mp4

Auto-captions

Transcribes your video locally with whisper.cpp and burns in styled captions. No API key needed. No cloud upload.

vidlet caption video.mp4                        # hormozi style
vidlet caption video.mp4 -s karaoke -c cyan      # karaoke, cyan
vidlet caption video.mp4 -s classic --srt my.srt  # your own SRT
Hormozi
Word-by-word color pop
Karaoke
Smooth color fill across words
Classic
White text, black outline
Minimal
Small text, background box

Jump cuts

Detects silence, cuts dead air, adds alternating punch-in zoom. The editing style behind every fast-paced YouTube video, in one command.

vidlet jumpcut video.mp4                # normal pace, 3% zoom
vidlet jumpcut video.mp4 --pace tight    # aggressive cuts
vidlet jumpcut video.mp4 --zoom 0        # no zoom, just cuts

All tools

vidlet caption <file>        # auto-captions (whisper.cpp)
vidlet jumpcut <file>        # auto-edit: cut silence + zoom
vidlet autocleanup <file>    # denoise + silence removal + compress
vidlet compress <file>       # H.264/HEVC compression
vidlet cleanvoice <file>     # neural voice denoising
vidlet removesilence <file>  # cut dead air
vidlet togif <file>          # optimized GIF
vidlet shrink <file>         # speed up for Shorts
vidlet extractaudio <file>   # pull audio track
vidlet mkv2mp4 <file>        # container conversion
vidlet thumb <file>          # set thumbnail
vidlet loop <file>           # seamless loop detection
vidlet optimize <file>       # compress Lottie/GIF

Every command supports -g (GUI) and -y (skip prompts).

MCP server

The same tools, driven by Claude Code, Claude Desktop or Cursor instead of typed by you. 27 tools over a local stdio server: your footage never leaves the machine.

npm i -g @spark-apps/vidlet
claude mcp add vidlet -- vidlet-mcp

For Claude Desktop or Cursor, add it to the app’s MCP config instead:

{
  "mcpServers": {
    "vidlet": { "command": "vidlet-mcp" }
  }
}

Added it while your agent was already running? Restart the session. An MCP tool list is negotiated once, when the session connects, so a server added mid-session stays invisible until then.

Example prompts

Once it is connected, just ask for what you want:

"Blur anything sensitive in demo.mp4 before I publish it"
"Preview the music beds, then score it with the calm one"
"Open that project in the VidLet editor"

A fuller one, spelling out the story order and the answers to the questions it would otherwise stop and ask:

Make a Short from these two screen recordings, modeling first
then rigging:
  ~/recordings/blender-modeling.mp4
  ~/recordings/blender-rigging.mp4
music: ~/music/synthwave-loop.mp3
Then give me the video url, the .vidlet source, and the render time.

Some tools stop and ask before rendering: which music bed, and whether the drafted narration is right. Your agent relays those questions to you and calls again with your answers, so nothing is encoded until you have approved the script.