A cloud-powered AI skill designed to automatically strip or repair burned-in and auto-generated captions from video files.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install caption-generator-downfall
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 caption-generator-downfall 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 Caption Generator Downfall?
Caption Generator Downfall is a specialized video processing agent built to streamline the workflow of content creators. By integrating this tool into your Openclaw Skills library, you can easily remove unwanted text overlays or fix broken auto-captions without manual frame-by-frame editing. The skill leverages cloud GPU acceleration to handle heavy lifting, providing a seamless experience for those who need to clean up video assets for republishing.
The system utilizes the Nemo Video API to perform sophisticated video compositing and platform-specific compression in the background. It is designed to work with common video formats like MP4 and MOV, ensuring that the output is high-quality and ready for use on any social media platform.
Caption Generator Downfall Use Cases
- Content creators who need to remove burned-in captions from existing videos for cross-platform posting.
- Social media managers looking to fix or disable inaccurate auto-generated subtitles.
- Marketing teams requiring clean video plates for adding new localized text or branding.
- Developers looking to automate video cleanup tasks within their Openclaw Skills powered agents.
How Caption Generator Downfall Works
- The skill authenticates with the backend using a NEMO_TOKEN or generates an anonymous session token for temporary use.
- A unique session is created to track the video editing project and maintain state across multiple requests.
- Users upload video files up to 500MB which are stored and processed on secure cloud GPU nodes.
- Editing commands are sent through a Server-Sent Events (SSE) stream, allowing for real-time interaction and status updates.
- The cloud render pipeline composites the video layers and applies H.264 compression at 1080p resolution.
- Upon completion, the skill provides a direct download link for the processed MP4 file.
Caption Generator Downfall Setup
To use this skill within your Openclaw Skills environment, ensure you have the required environment variables configured:
# Set your Nemo API token to enable high-priority rendering
export NEMO_TOKEN="your_api_token_here"
# Configuration paths are typically handled automatically at:
# ~/.config/nemovideo/
Caption Generator Downfall Data Schema & Taxonomy
The skill manages video data using a condensed JSON taxonomy to ensure fast communication between the agent and the cloud renderer:
| Property | Meaning |
|---|---|
t |
Track collection |
tt |
Track Type (0: Video, 1: Audio, 7: Text) |
sg |
Segments defining start/end times |
d |
Asset duration in milliseconds |
m |
Metadata and asset identification |
All exports default to 1080p MP4 using the H.264 codec to ensure maximum compatibility across devices.
Caption Generator Downfall Advanced Features
- Multi-asset batch processing within a single persistent session.
- Real-time session state polling to synchronize edits when stream responses are purely visual.
- Support for a wide range of media formats including MP4, MOV, AVI, WebM, and MKV.
- Intelligent intent classification to route natural language prompts to specific API actions like Export or State queries.
- Automatic anonymous token acquisition for users just starting out with Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Download Manager Skill
A comprehensive automation utility for managing qBittorrent and Transmission download clients through a unified interface.
HV-Analysis Deep Research Agent
A sophisticated research framework that combines longitudinal history with cross-sectional competitive analysis to generate exhaustive PDF reports.
Prediction Market Smoke Test
A minimal framework for sanity-checking prediction market questions and drafting quick trading theses.
Model Distill Master
A sophisticated workflow for distilling large language model capabilities into optimized smaller models like gemma-3-4b-it for edge deployment.
Cloudflare Global
A comprehensive AI agent skill for managing Cloudflare DNS records, zone settings, and tunnels using Global API Key authentication.
Wallet MCP: Multi-Chain Wallet Manager
A comprehensive multi-chain wallet management tool for Solana and EVM, enabling automated batch operations and secure local storage.
Featured*
