auto-remotion

Converts screen recordings into promotional videos using Remotion compositions, subtitles, voiceover, and BGM.

Updated Apr 17, 2026
One-click install
npx skills add https://github.com/16Miku/auto-remotion --skill auto-remotion-16miku
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-remotion
Source: https://github.com/16Miku/auto-remotion
Command: npx skills add https://github.com/16Miku/auto-remotion --skill auto-remotion-16miku

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires edge-tts, ffmpeg, remotion, elevenlabs-scribe.

What problem does it solve? Turning a long product demo recording into a short promotional video requires tedious manual editing, timeline planning, subtitle syncing, voiceover alignment, and rendering setup. This Skill provides a structured end-to-end workflow that converts raw screen recordings into polished Remotion-based promo videos. ## Core Features & Use Cases - Structured Editing Pipeline: Guides the full flow from goal alignment, storyboard (storyboard.json), and frame-level edit specs (edit-spec.json) to Remotion implementation, subtitles, Chinese voiceover via edge-tts, BGM, and final rendering. - Optional Automated Video Understanding: Uses ElevenLabs Scribe transcription with word-level timestamps and packed transcript files so an LLM can automatically identify valuable segments and generate storyboards. - Production Hard Rules: Includes 12 validated rules covering subtitle ordering, lossless concat, audio fades, word-boundary cuts, and timestamp padding to prevent silent failures. - Use Case: You have a 20-minute product demo recording and need a 60-second website promo video. The Skill walks you through identifying key moments, building the Remotion composition, generating synced Chinese voiceover with silence padding, adding BGM with dynamic volume, and rendering the final MP4. ## Quick Start Ask the AI to turn your screen recording into a 60-second promotional video using the auto-remotion workflow, starting with goal confirmation and storyboard planning.

Frequently Asked Questions about auto-remotion

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn a screen recording into a promotional video with Remotion?

Follow the staged workflow: confirm goals and duration, build a storyboard.json and edit-spec.json with frame-level clip timings, implement Remotion segments with trimmed Video components, then add subtitles, edge-tts voiceover, BGM, and render with npx remotion render.

How to fix voiceover and subtitle mismatch in Remotion videos?

The mismatch happens because TTS natural speech duration differs from target frame duration. Measure each generated segment's actual length, then use ffmpeg anullsrc to append silence padding so every voiceover segment matches its target timeline duration.

Can I use multiple source videos in one Remotion composition?

Yes. Define multiple entries in sourceVideos within edit-spec.json and reference each clip's src per segment. In the composition, pass the video identifier to the clip helper and load each file with staticFile from the public directory.

Does this workflow support automatic video understanding?

Yes, optionally. It uses ElevenLabs Scribe for word-level transcription, packs results into a takes_packed.md file, and lets an LLM read the transcript to automatically identify valuable segments and generate a storyboard JSON.

When should I not use this screen recording to promo workflow?

Do not use it when you have no source video, such as converting documents or slides directly to video, or when you need AI-generated visuals. Record a demo first with tools like OBS or Screen Studio, then apply this editing workflow.

Why does Remotion rendering fail with ENOSPC or frame range errors?

ENOSPC means insufficient disk space; clear the temp webpack bundle cache and ensure 5-10GB free. Frame range errors occur because ranges are left-closed right-open, so use --frames=0-{N-1} for a composition of N frames.