autocut-shorts

Automate short-form video clip creation from long videos with subtitles.

Updated Jan 30, 2026
One-click install
npx skills add https://github.com/akrindev/trimer-clip --skill autocut-shorts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autocut-shorts
Source: https://github.com/akrindev/trimer-clip/tree/main/skills/autocut-shorts
Command: npx skills add https://github.com/akrindev/trimer-clip --skill autocut-shorts

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires youtube-downloader, video-transcriber, speaker-diarization, scene-detector, laughter-detector, sentiment-analyzer, highlight-scanner, video-trimmer, portrait-resizer, subtitle-overlay, and includes scripts (resource) components.

What problem does it solve?

Long-form videos from podcasts, vlogs, tutorials, and streams are difficult to repurpose quickly for platforms like TikTok, YouTube Shorts, and Instagram Reels. Autocut Shorts automates the end-to-end workflow: downloading or ingesting the source video, transcribing audio, detecting highlights using transcript, laughter, sentiment, and scenes, trimming segments, resizing to 9:16, and adding subtitles for publish-ready clips.

Core Features & Use Cases

  • End-to-end automation: from source video to publish-ready 9:16 clips.
  • Multi-backend support: transcription and diarization options (Whisper, Gemini, Google, OpenAI; pyannote or Gemini for diarization).
  • Rich highlight detection: combines signals to score and select viral moments.
  • Platform-ready outputs: multiple clips with subtitles and metadata for editorial workflows.

Quick Start

Provide a long video file or URL and run the autocut script to generate multiple 9:16 clips ready for publication.

Frequently Asked Questions about autocut-shorts

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

FAQPage Schema
How do I automatically create short-form clips from long videos?

Automate short-form clip creation by ingesting long-form videos like podcasts, transcribing the audio, scoring highlights, and trimming segments into publish-ready 9:16 clips. This workflow handles download, diarization, and resizing end-to-end to generate multiple viral shorts automatically.

What is the best way to find viral moments in a podcast or vlog?

Finding viral moments involves combining multiple detection signals: transcript analysis, laughter detection, sentiment analysis, and scene detection. This multi-signal approach scores and selects the most engaging segments from long-form content to identify highlight clips accurately.

Can I use Whisper or Gemini for transcription and diarization when generating clips?

Yes, you can use multiple transcription backends including Whisper, Gemini, Google, and OpenAI, alongside diarization options like pyannote or Gemini. This multi-backend support ensures flexible audio processing when generating short-form video clips.

How do I resize and add subtitles to video clips for TikTok and Instagram Reels?

Resize videos to 9:16 and burn in subtitles by applying portrait resizing and subtitle overlay scripts. This process transforms trimmed highlight segments into platform-ready outputs for TikTok, YouTube Shorts, Instagram Reels, and Facebook Reels.

Does diarization help with trimming long videos into multiple short-form clips?

Diarization helps identify distinct speakers, ensuring trimmed segments maintain conversational context and coherence. Combined with highlight scanning and scene detection, it accurately isolates complete viral moments from long-form podcasts and vlogs for short-form clips.