Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Takes a long-form video file, transcribes it using OpenAI Whisper, identifies high-engagement segments suited for short-form platforms, cuts them with ffmpeg, and burns in animated subtitles in multiple styles.
Combines transcription, clip selection, cutting, subtitle styling, and format conversion into one pipeline instead of juggling Whisper, ffmpeg, and a caption editor separately.
When to use it
- Repurposing a podcast episode into TikTok clips
- Adding auto-captions to a recorded conference talk
- Finding the best 45-second moment from a 2-hour stream
- Converting a horizontal YouTube video to vertical Reels format
- Batch-cutting multiple viral moments from a single interview
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: here's my podcast recording, cut it for TikTok with word-highlight subtitles
- 1Transcribes the video using OpenAI Whisper API with word-level timestamps
- 2Analyzes the transcript and presents numbered clip options with time ranges, hook lines, and viral rationale
- 3User selects clips; agent cuts each segment to a TikTok-compatible mp4 using ffmpeg
- 4Generates word-highlight animated subtitles using the transcript timestamps and burns them into each clip
- 5Converts horizontal footage to 9:16 vertical format via center crop or blur-background padding
Final mp4 files saved to /home/ubuntu/clawd/clips/ with TikTok-compatible format, vertical aspect ratio, and animated subtitles
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.