muapi-ai-clipping

Extracts ranked vertical short clips from long videos via the muapi.ai ai-clipping API.

4.2k|481|Updated May 25, 2023
One-click install
npx skills add https://github.com/SamurAIGPT/Generative-Media-Skills --skill muapi-ai-clipping
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: muapi-ai-clipping
Source: https://github.com/SamurAIGPT/Generative-Media-Skills/tree/main/.opencode/skills/muapi-ai-clipping
Command: npx skills add https://github.com/SamurAIGPT/Generative-Media-Skills --skill muapi-ai-clipping

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires muapi-cli, and includes scripts (resource) components.

What problem does it solve?

Turning a long podcast, interview, or lecture into short-form clips for TikTok, Reels, or Shorts normally requires transcription, highlight selection, and manual cropping. This Skill replaces that pipeline with a single managed API call that returns ranked, face-tracked vertical clips.

Core Features & Use Cases

  • Automated highlight extraction: Server-side Whisper transcription and a virality framework (hooks, emotional peaks, opinion bombs, quotable lines) rank the best 30–75 second moments.
  • Face-tracked auto-crop: Outputs 9:16, 1:1, or 4:5 clips with no local ffmpeg, GPU, or LLM required.
  • Flexible output modes: Get hosted mp4 URLs per clip, or use --coords-only to retrieve just timestamps for your own renderer.
  • Use Case: Point the Skill at a 2-hour podcast URL and receive 5 scored vertical clips, each with a hook line, virality reason, and downloadable mp4.

Quick Start

Ask the agent to clip your video, for example: turn this podcast URL into five 9:16 short clips and show me the ranked results.

Frequently Asked Questions about muapi-ai-clipping

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

FAQPage Schema
How do I turn a long video into short clips for TikTok?

Run the ai-clipping script with your video URL and desired clip count, for example --video URL --num-clips 5. The API transcribes, ranks highlights by virality, and returns face-tracked 9:16 clips with hosted mp4 URLs.

How to extract highlight timestamps from a video without cropping?

Use the --coords-only flag to skip cropping and receive only start_time and end_time ranges for each ranked highlight. You can then render the clips locally with your own ffmpeg pipeline.

Does ai-clipping require local ffmpeg, Whisper, or a GPU?

No. All transcription, highlight ranking, deduplication, and face-tracked cropping run server-side on the muapi.ai endpoint. You only need muapi-cli installed and a valid MUAPI_API_KEY.

Can I use a local video file instead of a hosted URL?

Yes. Pass a local file path to --video and the script uploads it via muapi upload file before submitting the clipping job. YouTube URLs may also work if the backend supports them.

Why did I get fewer clips than I requested?

The API returns only highlights that survive virality ranking and overlap deduplication. If the source video has fewer rankable moments than num_clips, you receive what was found rather than padded results.