Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Security: Added requiredEnvVars metadata, comprehensive Security Notice explaining all capabilities (API keys, subprocess, cross-skill integration, file I/O), and code comments for legitimate automation features
What This Skill Does
Automates the YouTube content production pipeline: downloads a video, transcribes it with Whisper, generates SEO metadata with GPT-4, and creates AI thumbnails with consistent character styling. Supports Korean-language output and custom avatar images. Optionally integrates with the nano-banana-pro skill for image generation.
Combines transcription, metadata generation, and thumbnail creation into a single automated run instead of requiring separate tools for each step.
When to use it
- Generating subtitles for a recorded lecture before uploading
- Creating Korean SEO titles and tags for a gaming video
- Producing a branded thumbnail using your own photo as the avatar
- Batch-processing local video files into upload-ready packages
- Quickly generating a Pirate Lobster thumbnail for a prototype video
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: process this YouTube video and generate a thumbnail with my face
- 1Downloads the video from the provided YouTube URL using yt-dlp
- 2Transcribes the audio to an .srt subtitle file using OpenAI Whisper
- 3Sends transcript to GPT-4 to generate SEO-optimized title, description, and tags in Korean
- 4Generates a consistent character thumbnail using the user-supplied avatar image via nano-banana-pro
- 5Removes background from generated character and composes final Black & Gold thumbnail layout
Subtitle file, SEO metadata text, and a finished thumbnail image saved to the working directory
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.