ai-video-remix

Transform ShotAI-indexed footage into themed highlight reels as MP4 files.

4|Updated Mar 4, 2026
One-click install
npx skills add https://github.com/abu-ShotAI/ai-video-remix --skill ai-video-remix
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-video-remix
Source: https://github.com/abu-ShotAI/ai-video-remix/tree/main/skill
Command: npx skills add https://github.com/abu-ShotAI/ai-video-remix --skill ai-video-remix

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AI Video Remix removes the need to manually comb through footage and choreograph multi-step editing because it pairs a planning LLM with ShotAI semantic search and Remotion rendering to produce coherent remixes from local libraries.

Core Features & Use Cases

  • Semantic shot selection: Queries ShotAI MCP to locate the exact footage that matches LLM-crafted intent slots for travel vlogs, sports highlights, nature montages, or cyberpunk city slices.
  • Guided LLM orchestration: Agents parse intent, refine search queries, annotate clips, and decide on compositions, optionally falling back to heuristic mode when no API keys are available.
  • Deterministic rendering workflow: ffmpeg extracts trimmed clips, a local HTTP server feeds Remotion, and compiled BGM from yt-dlp or local files finishes the styled highlight reel.

Quick Start

Install the skill in Claude Code, configure SHOTAI_URL and SHOTAI_TOKEN, and then ask Claude to craft the remix you want from your library.

Frequently Asked Questions about ai-video-remix

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

FAQPage Schema
How do I automatically generate a travel vlog highlight reel from local footage?

Yes, you can create sports highlight reels and nature montages without manual editing by pairing a planning LLM with ShotAI semantic search to find exact footage matches and Remotion to render the final styled MP4 automatically.

What do I need to render Remotion videos from semantic search results?

To render Remotion videos from semantic search results, you need a running ShotAI MCP server, ffmpeg, nodejs, optional yt-dlp, and configured SHOTAI_URL and SHOTAI_TOKEN environment variables to plan, extract, and compile the final MP4.

How does semantic search find specific clips for a cyberpunk nightcut montage?

Semantic search finds specific clips for a cyberpunk nightcut montage by querying the ShotAI MCP server to locate exact footage that matches LLM-crafted intent slots, returning relevant shots from any sized local library without manual sorting.

Can I compile background music into highlight reels using yt-dlp and ffmpeg?

Yes, you can compile background music into highlight reels because the deterministic rendering workflow uses ffmpeg to extract trimmed clips and local HTTP servers to feed Remotion, finishing with BGM from yt-dlp or local files.

Does the video remix workflow function without API keys for ShotAI?

The video remix workflow functions without API keys by falling back to a heuristic mode where guided LLM agents parse intent, refine search queries, annotate clips, and decide on compositions using available local resources instead of the ShotAI API.