TimoteoAltobrando

AI Video Skill — AI skill for Claude Code

AI community

Make and edit videos with Claude Opus 5.5: a Claude Code skill for product films, ads, talking heads, explainers, montages and Reels/TikTok/Shorts, with Remotion, AI images and animation, ElevenLabs m.

How to install AI Video Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open TimoteoAltobrando/ai-video-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What AI Video Skill does

Make and edit videos with Claude Opus 5.5: a Claude Code skill for product films, ads, talking heads, explainers, montages and Reels/TikTok/Shorts, with Remotion, AI images and animation, ElevenLabs music and ffmpeg.

Alternatives in AI

  • Qwen Audio Agent — A realtime voice runtime that keeps Agents talking, working, and present 2.3k ★
  • Gsap Skills — Official AI skills for GSAP 1.7k ★
  • Spine Animation AI — AI-powered Claude skill for Spine 2D skeletal animation — auto-rig, animate, and preview characters 382 ★

README

ai-video-skill

**Make and edit videos with Claude Opus 5.5.** A [Claude Code](https://claude.com/claude-code) skill, written and tested with Opus 5.5, that takes a video from the first idea, or from your own footage, to the finished files: product and launch films, ads, talking-head and founder videos, interviews, explainers and tutorials, event and trip montages, podcast clips, Reels/TikTok/Shorts, animated logos, music videos, and changes to a finished video (a shorter cut, a vertical version, subtitles, a translation, new music).

Remotion is the backbone; images come from Codex (`image_gen`), AI animation from Seedance (via OpenRouter), music and sound effects from ElevenLabs, footage and speech are handled with ffmpeg and whisper. Claude first asks what kind of video it is and everything that kind needs, then proposes the direction, gets the script (or the paper edit of the footage) approved piece by piece and produces the video, round after round, until delivery.

The references hold what makes a video work and what to avoid: direction, music, the person on camera, editing footage, delivery for each platform.

What it does

  1. Intake: where we start from (from scratch, footage, both, a finished video), the kind of video, the platforms, the length; then the core (goal, audience, materials, voice, words, music, AI or not, rights and consent, budget) and the questions specific to that kind. A one-screen brief to confirm before any work.
  2. Tools and keys: only those the chosen path needs, stored safely, with spending caps.
  3. Analysis: brand, product, footage (inventory, contact sheets, transcripts), references.
  4. Direction: story ideas with images and a script with one image per piece, or, for footage, the selects and a paper edit with one frame per shot; hooks for short verticals.
  5. The person on camera (if any): the room, the light, the voice, the graphics around them; a shooting guide if it is not shot yet.
  6. **Product