Rendervid MCP Skill
SkillMediaGenerate videos and images from JSON templates using AI-powered video rendering
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Rendervid MCP Skill skill
What this skill tells your AI
The instructions your AI receives, as published by qualityunit/rendervid in skills/SKILL.md and read by ahel’s review.
Generate professional videos and images from JSON templates using natural language. Rendervid provides a declarative, AI-friendly way to create animated content for social media, marketing, data visualization, and more.
Overview
Rendervid is a Model Context Protocol (MCP) server that enables AI agents to:
- Generate videos in multiple formats (MP4, WebM, MOV, GIF)
- Create static images (PNG, JPEG, WebP)
- Access 50+ ready-to-use templates
- Validate templates before rendering
- Discover all available capabilities
Key Features:
- Declarative JSON Templates - Define videos/images as simple JSON structures
- 40+ Animation Presets - Entrance, exit, and emphasis animations with easing functions
- Rich Layer Types - Text, images, video, shapes, audio, Lottie, custom React components
- 50+ Examples - Ready-to-use templates across 10+ categories
- Type-Safe - Full TypeScript with Zod validation
- Cross-Platform - Works in Node.js and browsers
When to Use This Skill
Use Rendervid when you need to:
Content Creation
- Generate social media content (Instagram stories, TikTok videos, YouTube thumbnails)
- Create marketing videos (product showcases, sale announcements, testimonials)
- Build data visualizations (animated charts, graphs, dashboards)
- Produce educational content (course intros, lesson titles)
Automation
- Batch generate videos from data
- Create personalized video content
- Generate thumbnails for video content
- Build video generation workflows
Prototyping
- Quickly test video concepts
- Create animated mockups
- Generate placeholder content
- Explore animation styles
AI-Driven Content
- Let AI agents create videos from text descriptions
- Generate videos based on user input
- Create dynamic content based on data
- Build video generation chatbots
Available Tools
The Rendervid MCP server provides 6 tools:
1. render_video
Generate video files from JSON templates.
Common uses:
- Social media videos
- Marketing content
- Animated explainers
- Data visualizations
Example prompt:
"Create a 5-second Instagram story with text 'Summer Sale' and a blue gradient background"
2. render_image
Generate static images or single frames.
Common uses:
- Thumbnails
- Social media images
- Preview frames
- Static graphics
Example prompt:
"Generate a YouTube thumbnail with title 'Best Practices 2024'"
3. validate_template
Validate template JSON before rendering.
Common uses:
- Check template syntax
- Catch errors early
- Verify AI-generated templates
- Debug template issues
Example prompt:
"Validate this template: [paste JSON]"
4. get_capabilities
Discover all available features.
Common uses:
- Learn what's possible
- Find available animations
- Check supported formats
- Understand layer types
Example prompt:
"What animation presets are available?"
5. list_examples
Browse 50+ example templates.
Common uses:
- Find templates by category
- Discover use cases
- Get inspiration
- Learn template structure
Example prompt:
"Show me all social media templates"
6. get_example
Load a specific example template.
Common uses:
- Use ready-made templates
- Start from examples
- Learn template patterns
- Customize existing templates
Example prompt:
"Load the Instagram story template"
Template Structure
Rendervid templates are JSON objects with three main sections:
{
"name": "My Video",
"output": {
"type": "video",
"width": 1920,
"height": 1080,
"fps": 30,
"duration": 5
},
"inputs": [
{
"key": "title",
"type": "string",
"label": "Title",
"default": "Hello"
}
],
"composition": {
"scenes": [
{
"id": "main",
"startFrame": 0,
"endFrame": 150,
"backgroundColor": "#1a1a2e",
"layers": [
{
"id": "title",
"type": "text",
"position": { "x": 960, "y": 540 },
"size": { "width": 1600, "height": 200 },
"inputKey": "title",
"props": {
"fontSize": 120,
"color": "#ffffff",
"textAlign": "center"
},
"animations": [
{
"type": "entrance",
"effect": "fadeInUp",
"duration": 30,
"easing": "easeOut"
}
]
}
]
}
]
}
}
Key Concepts
Output Configuration:
- Defines video/image dimensions, FPS, duration
- Supports multiple formats and codecs
Dynamic Inputs:
- Variables that can be customized per render
- Support strings, numbers, booleans, colors, images, videos
Composition:
- Scenes with frame ranges
- Layers with types, positions, props
- Animations with presets and easing
Common Patterns
Pattern 1: Text Animation
User: "Create a video with animated text"
AI flow:
1. get_capabilities() to understand animations
2. render_video() with fadeIn animation
Pattern 2: Social Media Content
User: "Make an Instagram story for a product launch"
AI flow:
1. list_examples({ category: "social-media" })
2. get_example({ examplePath: "social-media/instagram-story" })
3. render_video() with custom inputs
Pattern 3: Data Visualization
User: "Create an animated bar chart showing Q1 sales"
AI flow:
1. get_example({ examplePath: "data-visualization/animated-bar-chart" })
2. Modify template with data
3. render_video()
Pattern 4: Batch Generation
User: "Create 10 videos with different titles"
AI flow:
1. Load template once
2. Call render_video() 10 times with different inputs
Pattern 5: Template Validation
User: "Check if my custom template is valid"
AI flow:
1. validate_template() to check structure
2. Fix any errors
3. render_video() to generate
Special Topics
Animations
Rendervid supports 40+ animation presets organized by category:
Entrance Animations:
- fadeIn, fadeInUp, fadeInDown, fadeInLeft, fadeInRight
- slideInUp, slideInDown, slideInLeft, slideInRight
- scaleIn, zoomIn, rotateIn, bounceIn, flipInX, flipInY
Exit Animations:
- fadeOut, fadeOutUp, fadeOutDown, fadeOutLeft, fadeOutRight
- slideOutUp, slideOutDown, slideOutLeft, slideOutRight
- scaleOut, zoomOut, rotateOut, bounceOut
Emphasis Animations:
- pulse, shake, bounce, swing, wobble, flash, rubberBand
- heartbeat, float, spin, tada, jello, flip
Easing Functions (30+):
- Linear: linear
- Ease: ease, easeIn, easeOut, easeInOut
- Cubic: easeInQuad, easeOutQuad, easeInOutQuad
- And more: Cubic, Quart, Quint, Sine, Expo, Circ, Back, Elastic, Bounce
Layer Types
Text Layers:
- Typography controls (font, size, weight, style)
- Alignment and spacing
- Color and effects
- Google Fonts support
Image Layers:
- Fit options (cover, contain, fill)
- Scaling and positioning
- Filters and blend modes
Video Layers:
- Playback controls
- Timing and trimming
- Audio control
Shape Layers:
- Rectangles, ellipses, polygons, stars, paths
- Fill and stroke
- Border radius and effects
Audio Layers:
- Background music
- Sound effects
- Volume and fade controls
Lottie Layers:
- Animated vector graphics
- JSON animation files
Custom Layers:
- React components
- Full CSS/Tailwind support
- Advanced use cases
Output Formats
Video Formats:
- MP4 (H.264, H.265)
- WebM (VP8, VP9, AV1)
- MOV (ProRes, H.264)
- GIF
Image Formats:
- PNG (lossless)
- JPEG (lossy, quality control)
- WebP (modern, efficient)
Quality Presets:
- draft: Fast rendering, lower quality
- standard: Balanced quality/speed
- high: High quality, slower rendering
- lossless: Maximum quality, large files
Template Categories
Getting Started:
- Hello World
- First Video
- First Image
Social Media:
- Instagram Story (1080x1920)
- Instagram Post (1080x1080)
- TikTok Video (1080x1920)
- YouTube Thumbnail (1280x720)
- Twitter Card (1200x630)
- LinkedIn Banner (1584x396)
Marketing:
- Product Showcase
- Sale Announcement
- Testimonial Video
- Before & After
- Logo Reveal
Data Visualization:
- Animated Bar Chart
- Line Graph
- Pie Chart
- Counter Animation
- Progress Dashboard
Business:
- Quote Card
- Stat Highlight
- Feature Announcement
- Team Introduction
- Company Update
Education:
- Lesson Title
- Quiz Question
- Course Intro
- Study Tip
- Progress Tracker
Celebrations:
- Birthday Card
- Holiday Greeting
- Congratulations
- Thank You
- Milestone
E-commerce:
- Product Launch
- Flash Sale
- New Arrival
- Customer Review
- Price Drop
Event:
- Event Announcement
- Countdown Timer
- Registration
- Thank You (Attendees)
- Highlight Reel
Podcast:
- Episode Intro
- Quote Card
- Guest Introduction
- Coming Soon
- Subscribe Reminder
Real Estate:
- Property Showcase
- Open House
- Sold Announcement
- Agent Introduction
- Virtual Tour Intro
Best Practices
Template Design
-
Keep it Simple
- Start with examples
- Build complexity gradually
- Test frequently
-
Use Dynamic Inputs
- Make values customizable
- Provide sensible defaults
- Document input purposes
-
Optimize Frame Ranges
- Minimize total frames for faster rendering
- Use appropriate FPS (24-30 for most content)
- Consider file size vs quality tradeoffs
-
Test Animations
- Preview before full render
- Check timing and easing
- Verify layer stacking
Performance
-
Draft Quality for Iteration
- Use draft quality during development
- Switch to high quality for finals
- Consider file size requirements
-
Batch Rendering
- Reuse templates when possible
- Queue multiple renders
- Monitor resource usage
-
Asset Optimization
- Compress images before use
- Use appropriate video codecs
- Optimize Lottie animations
AI Integration
-
Clear Descriptions
- Describe intent clearly
- Specify dimensions and duration
- Mention animation preferences
-
Validate Templates
- Always validate AI-generated templates
- Check for common errors
- Test with sample data
-
Iterate Progressively
- Start simple, add complexity
- Test each change
- Use examples as starting points
Example Workflows
Workflow 1: Quick Social Media Post
User: "Create an Instagram story announcing a flash sale"
AI Agent:
1. list_examples({ category: "social-media" })
2. get_example({ examplePath: "social-media/instagram-story" })
3. render_video({
template: {...},
inputs: {
title: "Flash Sale",
subtitle: "50% Off Everything",
backgroundColor: "#FF6B6B"
},
format: "mp4",
quality: "high"
})
Result: Instagram-ready 1080x1920 video
Workflow 2: Data Visualization
User: "Create an animated chart showing Q1 revenue growth"
AI Agent:
1. get_example({ examplePath: "data-visualization/animated-bar-chart" })
2. Modify template with revenue data
3. validate_template({ template })
4. render_video({
template,
inputs: { data: [...], title: "Q1 Revenue" },
format: "mp4",
quality: "high"
})
Result: Animated bar chart video
Workflow 3: Custom Template Creation
User: "Create a custom video with my brand colors and logo"
AI Agent:
1. get_capabilities() to understand options
2. Start with example template
3. Customize colors, fonts, layout
4. validate_template() to check
5. render_video() with draft quality
6. Iterate based on preview
7. Final render with high quality
Result: Custom branded video
Workflow 4: Batch Video Generation
User: "Create welcome videos for 100 new users"
AI Agent:
1. get_example({ examplePath: "business/team-introduction" })
2. Loop through user list:
- render_video({
template,
inputs: { name: user.name },
outputPath: `welcome-${user.id}.mp4`
})
Result: 100 personalized welcome videos
Troubleshooting
Common Issues
Issue: Server not connecting
- Check absolute path in MCP config
- Verify Node.js version (20+)
- Check server builds:
pnpm build - Review logs in AI editor
Issue: FFmpeg not found
# Install FFmpeg
brew install ffmpeg # macOS
sudo apt install ffmpeg # Ubuntu
choco install ffmpeg # Windows
Issue: Template validation fails
- Check JSON syntax
- Verify required fields
- Review error messages
- Use validate_template tool
Issue: Slow rendering
- Use draft quality for iteration
- Reduce video duration/FPS
- Optimize asset sizes
- Check system resources
Issue: Node version errors
- Update to Node.js 20+
- Use full Node.js path in config
- Verify with:
node --version
Getting Help
- Documentation: MCP Server README
- Examples: Browse
/examples/directory - Issues: GitHub Issues
- Discussions: GitHub Discussions
Resources
Documentation
Templates
External Links
Version History
0.1.0 (Initial Release)
- 6 MCP tools (render_video, render_image, validate_template, get_capabilities, list_examples, get_example)
- 50+ example templates across 10 categories
- 40+ animation presets with 30+ easing functions
- Support for MP4, WebM, MOV, GIF, PNG, JPEG, WebP
- Full TypeScript with Zod validation
- Stdio transport for maximum compatibility
- Comprehensive documentation
License
FlowHunt Attribution License - Free for commercial and personal use with attribution
Contributing
Contributions welcome! See CONTRIBUTING.md for guidelines.
Credits
Developed by QualityUnit
Built with:
- TypeScript
- Model Context Protocol SDK
- Zod for validation
- Puppeteer for rendering
- FFmpeg for video encoding
Signals
- GitHub stars
- 63
- Forks
- 12
- Last commit
- Aug 2026
Advanced
- Catalog kind
- skill
- Gateway key
rendervid- Source
- github.com/qualityunit/rendervid