create_projectLABSCO SUMMARY
A video built in pieces, not in one prompt.
Holding the work as a project with scenes is what makes it finishable: each shot, voice line and track is generated separately, and the running cost is visible before the next call rather than after.
What it is
A video production server built on FAL AI models. A project holds scenes; images are generated from text or from another image, clips from images, and speech and music generated to sit under them. Cost is estimated and tracked as you work.
What you get
- A project created, and scenes added to it one at a time
- Images generated from a text prompt, or from an image you already have
- Video generated from an image
- Speech and background music generated for the cut
- Your projects listed, so an unfinished one can be picked up
- Platform presets for YouTube, TikTok and Instagram, with running cost estimates
Requirements
Python 3.11 or newer, FFmpeg installed on the machine, uv, and a FAL AI key in the environment. YouTube upload goes through OAuth2, and the optional YouTube search features take a Google API key.
Setup effort
Build from source — git clone https://github.com/h2a-dev/video-gen-mcp-monolithic
USE CASES
01
Making a short video from an idea
Open a project, generate the frame from a prompt, and turn that frame into motion — the ordinary path from nothing to a clip.
02
Giving a cut a voice track
Speech is generated per scene rather than over the whole video, so a line can be redone without re-rendering everything around it.
03
Scoring a finished sequence
Music is generated last, against a project whose scenes already exist, so the length and mood match what is actually on screen.
04
Building on art you already have
An existing image becomes the starting frame — a product shot, a logo, a still — instead of describing it and hoping.
05
Picking up an unfinished project
Projects are listed and reopened, which matters when generation is slow and the work spans more than one sitting.
15 tools
Listed in project docs. Your client may show a different list.
add_scenelist_projectsgenerate_image_from_textgenerate_image_from_imagegenerate_video_from_imagegenerate_musicgenerate_speechdownload_assetsadd_audio_trackassemble_videoanalyze_scriptsuggest_scenesupload_image_fileget_server_infoBEFORE YOU INSTALL
What to know first
Tools you get
15 tools named in the project's documentation.
Requires
- Python 3.11+
- FALAI_API_KEY in your environment
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
TERMINAL
git clone https://github.com/h2a-dev/video-gen-mcp-monolithic
cd video-gen-mcp-monolithic
uv syncMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"video-agent": {
"command": "uv",
"args": [
"--directory",
"<YOUR_PATH_TO_VIDEO_GEN_MCP_MONOLITHIC>",
"run",
"python",
"main.py"
],
"env": {
"FALAI_API_KEY": "<YOUR_FALAI_API_KEY>"
}
}
}
}
