An AI-powered pipeline to reverse-engineer viral videos into replicable Seedance 2.0 prompts with intelligent material replacement.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install viral-video-replicator
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 viral-video-replicator 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 Viral Video Replicator?
The Viral Video Replicator is a sophisticated technical skill designed to decompose reference videos into structured data for creative replication. By utilizing a pipeline of local tools and cloud-based AI, it allows developers and creators to extract the core 'DNA' of viral content. This skill is a core part of the Openclaw Skills ecosystem, enabling users to transform a source video into a high-fidelity Seedance 2.0 prompt that can preserve or replace specific elements like the actor's face or the clothing being worn.
At its heart, the skill bridges the gap between raw video files and structured AI prompts. It handles the heavy lifting of frame extraction, audio transcription, and visual analysis, providing a turnkey solution for content creators who need to scale their production based on successful social media trends or competitor analysis.
Viral Video Replicator Use Cases
- Reverse-engineering viral fashion videos to extract clothing details and scene logic.
- Replacing garments in existing video content for e-commerce marketing while maintaining the original model's movements.
- Swapping actors in high-performing video templates to maintain brand consistency.
- Batch processing social media trends to generate a library of actionable production prompts.
- Analyzing competitor content to understand engagement triggers like lighting, camera angles, and dialogue scripts.
How Viral Video Replicator Works
- Environment Verification: The skill first checks for local FFmpeg and ffprobe installations to ensure media processing capabilities.
- Parallel Media Processing: It concurrently extracts frame grids at 3fps and the audio track to minimize processing time.
- ASR Transcription: If dialogue is detected, the audio is uploaded to TOS for automated speech-to-text transcription.
- Visual LLM Analysis: The extracted frame grids are analyzed by a Vision LLM to identify person attributes, clothing textures, scene context, and timestamped actions.
- Replacement Determination: Based on user-provided materials (like a new garment or face image), the skill selects one of four modes: Clone, Face Swap, Outfit Swap, or Full Swap.
- Prompt Assembly: A final Seedance 2.0 prompt is generated, integrating @image tags and specific technical constraints to ensure high-fidelity output.
- Validation: The system validates the JSON schema and prompt syntax before delivering the final assets to the user.
Viral Video Replicator Setup
To use this skill within the Openclaw Skills framework, you must first ensure your local environment is prepared.
# Install FFmpeg (Required for frame and audio extraction)
# macOS
brew install ffmpeg
# Linux
sudo apt update && sudo apt install ffmpeg
# Verify installation
ffmpeg -version && ffprobe -version
Next, configure your API credentials. The skill requires an ARK_API_KEY for Vision analysis. Optional ASR and TOS credentials are required if you wish to process video dialogue.
Viral Video Replicator Data Schema & Taxonomy
The skill organizes its analysis into a structured format compatible with the wider Openclaw Skills metadata standards:
| Category | Description |
|---|---|
person |
Maps facial features, hair style, skin tone, and body build. |
clothing |
Details garment type, fabric texture, fit, and specific design elements. |
scene |
Identifies the physical location and primary lighting sources. |
actions |
A chronological log of movements (e.g., 0-2s: waving). |
dialogue |
The full transcribed text of the video's audio track. |
Viral Video Replicator Advanced Features
- Support for 4 distinct replacement modes (Clone, Face Swap, Outfit Swap, Full Swap) for granular control over output.
- Multi-agent compatibility allowing for sequential or batch processing of large video libraries.
- Graceful degradation logic that provides visual-only analysis if ASR or audio components fail.
- Precise 'Exact' analysis mode for high-fidelity 1:1 replication and 'Rewrite' mode for thematic logic extraction.
- Automatic SOP (Standard Operating Procedure) generation for manual upload steps in video creation platforms.
SKILL.md
Loading
Related Openclaw Skills
MCP Server Pack
A curated bundle of production-ready Model Context Protocol (MCP) servers offering secure filesystem access, database queries, and third-party integrations.
Aliyun CCC Manage
A specialized technical skill for managing Alibaba Cloud Call Center resources through automated OpenAPI and SDK workflows.
Alibaba Cloud Chatbot Manager
A specialized tool for managing Alibaba Cloud Chatbot (beebot) resources through automated OpenAPI and SDK integrations.
JS X Monitor
An automated monitoring tool for X.com that tracks specific accounts and sends instant notifications for new tweets.
ComfyUI Client
A robust Node.js client for automating image and video generation workflows through the ComfyUI API.
session-sync-cloud
A professional-grade cloud synchronization skill that provides encrypted, versioned backups for AI agent memory files across S3-compatible storage.
Featured*
