A skill that generates AI-powered video messages featuring lip-synced avatars to provide human-like async updates.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install send-video-message
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install send-video-message using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Send Video Message?
The Send Video Message skill allows AI agents to communicate using personalized AI avatars that speak any provided text with natural motion and lip-syncing. Powered by the Runway Character API, this tool transforms dry text updates into engaging video content, making it ideal for situations where a face and voice add more value than a wall of text. It integrates seamlessly into the agent workflow to provide a more empathetic and professional touch to user interactions.
By leveraging Openclaw Skills, developers can automate the creation of these videos for various platforms. The skill supports custom avatars—including the signature Openclaw Lobster—as well as text-to-avatar generation. Whether you are sending a quick status update or a detailed technical walkthrough, this skill handles the rendering, voice synthesis, and platform-specific cropping automatically.
Send Video Message Use Cases
- Deploy Notifications: Provide a friendly face to confirm successful production deploys and performance improvements.
- Code Review Walkthroughs: Explain complex PR feedback visually to reduce friction and improve understanding.
- Incident Summaries: Deliver calm, clear recaps of system incidents and the steps taken to resolve them.
- Async Standups: Send automated daily or weekly progress reports to keep the team aligned without a live meeting.
- Onboarding & Education: Explain codebase architecture or new features using a consistent brand avatar.
How Send Video Message Works
- Avatar Initialization: The agent checks for an existing avatar profile in the local configuration; if none exists, it creates one using a reference image or text description.
- Script Processing: The agent receives a text string and selects an appropriate TTS (Text-to-Speech) voice based on the avatar's persona.
- Runway API Call: The text and avatar ID are sent to Runway's Character API to generate a lip-synced video.
- Post-Processing: If requested, the skill uses ffmpeg to crop the video into square (1:1) or vertical (9:16) formats for mobile optimization.
- Delivery: The final MP4 file is saved to a temporary directory and emitted via a MEDIA tag for the Openclaw environment to attach to the chat.
Send Video Message Setup
Prerequisites
- A Runway API Key from dev.runwayml.com.
- The
uvpackage manager installed. ffmpeg(optional, for cropping support).
Installation
# Set your API environment variable
export RUNWAY_API_SECRET=your_key
# One-time setup to create your custom avatar (e.g., the Openclaw Lobster)
uv run scripts/setup_avatar.py \
--name "Mochi" \
--image-url "https://runway-static-assets.s3.us-east-1.amazonaws.com/calliope-demo/agent-avatars-presets/openclaw-lobster2.jpg"
Send Video Message Data Schema & Taxonomy
The skill manages persistent data and output files using the following structure:
| Type | Location / Format | Description |
|---|---|---|
| Configuration | ~/.openclaw/runway-avatar.json |
Stores the unique avatar_id for reuse across sessions. |
| Video Output | /tmp/runway-avatar-YYYY-MM-DD-HH-MM-SS.mp4 |
The rendered video file ready for delivery. |
| Metadata | JSON Environment Variables | Uses RUNWAY_API_SECRET for authentication and SEND_VIDEO_MESSAGE_SQUARE/VERTICAL for formatting. |
Send Video Message Advanced Features
- Text-to-Avatar Generation: Use descriptive prompts to have Runway generate a completely unique face for your agent.
- Mobile-Specific Cropping: Intelligent center-cropping for Telegram (square) and WhatsApp/iMessage (vertical) using ffmpeg hooks.
- Extensive Voice Library: Choose from over 20 distinct TTS voices with varying genders, pitches, and emotional tones.
- Fallback Presets: Quickly switch to built-in Runway presets like
game-characterif custom avatar creation is not feasible. - Async Media Attachment: Automatically signals the host environment to upload and display the video via the MEDIA: protocol.
SKILL.md
Loading
Related Openclaw Skills
Zotero Scholar
A professional research assistant that synchronizes paper metadata, links, and AI-generated insights directly into your Zotero library.
YesCaptcha
YesCaptcha provides a robust API-based service for automating the recognition and solving of various CAPTCHA types during web automation.
The Ye Protocol
A specialized AI framework that transforms standard model outputs into the visionary, high-confidence rhetorical style of Kanye West.
Daily News Reporter
An AI-assisted toolkit for generating professional international news reports using structured multi-source verification and PDF automation.
Antigravity IDE Integration
A revolutionary agent-first development environment by Google designed to place AI agents at the center of the software creation lifecycle.
Social Video Content
A specialized developer tool for programmatically generating platform-optimized social media videos using React and Remotion.
Featured*
