What does this skill do?

The AI Video Editing Skill transforms long, unorganized recordings—screen recordings, vlogs, interviews, or demos—into polished, publish-ready videos. It doesn’t generate video from scratch: it combines AI-powered planning with FFmpeg’s deterministic cutting, Remotion’s programmable compositing, and assets generated by ElevenLabs and fal.ai to quickly and consistently compress, structure, and enhance existing footage.

Cutting and Organizing Recordings
It transcribes hours of footage, identifies the most compelling segments, and generates FFmpeg commands to automatically cut and concatenate them.
Cross-platform reframing
Convert a 16:9 video to 9:16 for TikTok/Reels or 1:1 for Instagram with a single FFmpeg command.
Overlays and Titles with Remotion
Add animated titles, graphics, and reusable templates using programmable composition in code.
Generated voice-over and music
Generate only what you need with ElevenLabs and fal.ai: voice-overs, music, sound effects, or b-roll on demand.

Usage examples

✂️ Organizing Materials
Here is the transcript of a 4-hour recording. Identify the 8 most compelling segments for a 24-minute vlog and give me the FFmpeg commands to cut each one.
📱 Vertical cropping
Convert this 16:9 video to a 9:16 vertical format for TikTok while keeping the subject centered.
🔇 Silence Detection
Analyze this video and identify all pauses longer than 2 seconds so you can eliminate the dead air.
🎨 Overlay with Remotion
Create a Remotion component with an animated title that appears at frame 30 and lasts for 90 frames over the intro.mp4 segment.

Features

6-layer pipeline Each layer has a specific function: capture, AI-powered organization, editing with FFmpeg, compositing with Remotion, generated assets, and final finishing.
Deterministic Cuts with FFmpeg Split, trim, concatenate, and normalize audio in an automated and fully reproducible manner.
Programmable Composition with Remotion Add overlays, titles, animated graphics, and reusable templates using React code.
On-Demand Assets Generate voiceovers with ElevenLabs and music or effects with fal.ai only for the missing parts—never the entire video.
Automatic scene and silence detection Automatically identifies downtime and scene changes to speed up segment selection.

Frequently asked questions

No. The goal is to compress, organize, and enhance existing real-world recordings. The AI plans and generates commands; FFmpeg cuts the footage; Remotion composes it; ElevenLabs and fal.ai generate only specific assets.
FFmpeg is required for editing. If you want programmable compositing, you'll need Remotion (Node.js). For generated assets, you'll need API keys from ElevenLabs and fal.ai. The final touches are done in Descript or CapCut.
Yes. If you're going to reuse a format more than once, turn it into a reusable Remotion component and just change the input segments.
Yes. The skill includes commands to reframe to 9:16, 1:1, and 16:9 so you can export platform-specific versions from a single recording.
AI Video Editing — Automated Editing Pipeline with Claude AI

¿Prefieres escuchar el contenido? Genera la narración de audio con un clic.