🎬 Medeo Video Generator
Generate AI-powered videos from text. Tell your AI assistant "make me a video about XX" and it will automatically create the video using the Medeo platform, sending you the result in a few minutes.
Installation
Option A: Install via OpenClaw (Recommended)
Tell your AI assistant:
Install the medeo-video skill from https://github.com/one2x-ai/medeo-video-skill
Option B: Manual Install
cd ~/.agents/skills/
git clone https://github.com/one2x-ai/medeo-video-skill.git medeo-video
openclaw gateway restart
Setup (One-Time)
The first time you use the skill, your assistant will prompt you to set up an API key:
- Go to 👉 https://medeo.app/dev/apikey
- No account? The page will guide you through registration
- Copy your API key (starts with
mk_) - Send it to your assistant — it will handle the rest
That's it. One-time setup, works forever.
Usage
Text to Video
Just describe what you want:
Make me a video about the coffee brewing process with a cozy vibe
Create a 30-second product launch video for new running shoes
Image + Text to Video
Send an image (or paste a URL) along with your description:
Use this image to create a product promo video
[send your image]
Make a video using this photo: https://example.com/photo.jpg
Your assistant will automatically upload your image and ensure the video uses your provided assets.
Vertical / Short-Form Video
Want TikTok / Reels / Shorts format? Just say so:
Make a vertical short video of a sunset timelapse
What You Can Do
| What You Say | What Happens |
|---|---|
| "Make me a XX video" | Full auto video generation |
| "Use this image to make a video" + image | Uses your image as assets |
| "Make it vertical" / "9:16" | Portrait short-form video |
| "What video templates are available?" | Browse preset recipes |
| "Is my last video done?" | Check recent job status |
| "Show my video history" | View past generations |
FAQ
How long does generation take?
Typically 5–15 minutes. Complex scripts may take longer. Generation runs in the background — your assistant will send you the video when it's ready.
What image/video formats are supported?
.jpg, .png, .webp, .mp4, .mov, .gif
What if I run out of credits?
Your assistant will let you know and provide a link to top up. After recharging, just say "credits topped up, please retry."
Which chat platforms are supported?
All platforms supported by OpenClaw: Feishu, Telegram, Discord, WhatsApp, Signal, and more. The video will be delivered directly to your chat.
Technical Details
- Requirements: Python 3.6+ with
requestslibrary (pip install requests) - Data Storage:
~/.openclaw/workspace/medeo-video/(config + job history) - API Docs: https://docs.prd.medeo.app/
Feishu Video Delivery
The optional feishu_send_video.py script reads Feishu app credentials (appId + appSecret) from your local ~/.openclaw/openclaw.json to obtain a tenant_access_token for sending videos. No credentials are uploaded or transmitted to third parties — they are only used locally to call the Feishu Open API. If you don't use Feishu, this script is not invoked.
Links
MIT License