Auto-generate viral 9:16 YouTube Shorts (or TikTok / Reels clips) from a long-form video. Thin platform-aware wrapper around the AI Clipping skill — picks sensible defaults for short-form social platforms (9:16, 30–60s sweet spot) and delegates the actual highlight extraction + crop to muapi.ai's `/ai-clipping` endpoint.
This skill auto-generates vertical short-form video clips (YouTube Shorts, TikTok, and Instagram Reels or Feed) from a long-form source video. It solves the manual, time-consuming task of finding the most engaging moments in a long video and cropping them to the right aspect ratio for each platform. It is a thin, platform-aware preset over an underlying AI Clipping primitive: it chooses sensible defaults (aspect ratio and clip count) for the target platform and delegates the heavy lifting to muapi.ai's managed /ai-clipping endpoint.
All processing runs server-side, so no local ffmpeg, Python, Whisper, or LLM keys are needed. A single shell script resolves the source (uploading local files to the muapi CDN when necessary), applies platform defaults, calls the clipping endpoint, polls until complete, and prints a ranked summary that can optionally download or open the clips. The remote pipeline transcribes the audio, ranks highlights through a virality framework (hook moments, emotional peaks, opinion bombs, revelations, conflict, quotable lines, story peaks, practical value), dedupes overlapping candidates, selects the top N, and produces face-tracked vertical crops. Each returned clip includes a 0-100 score, an opening hook line, a one-sentence rationale, a time range, and a clip URL. It supports overrides, batch processing from a URL list, and asynchronous submit-and-poll.
It targets content creators, social media managers, and marketers who want to repurpose long videos into ranked short clips quickly. Prerequisites are minimal: the muapi CLI installed and authenticated and a MUAPI_API_KEY. The skill uses that key through the official CLI, uploads source media to the vendor's CDN for processing, and includes explicit guidance not to fabricate keys, making it a straightforward read-and-call API wrapper with standard credential handling.
Ranked vertical short clips from a long video, each with a score (0-100), an opening hook line, a virality rationale, a time range, and a clip URL, tuned to the target platform's aspect ratio and clip count.
The muapi CLI installed and authenticated (muapi auth configure) and a MUAPI_API_KEY. No local ffmpeg, Python, Whisper, or LLM keys are needed because processing runs server-side.
YouTube Shorts, TikTok, and Instagram Reels (all 9:16) and Instagram Feed (1:1), each with default clip counts you can override with --aspect-ratio and --num-clips.
The source is resolved from a YouTube URL, a hosted mp4 URL, or a local file; local files are uploaded to muapi's CDN, and the clipping pipeline runs on muapi.ai's servers.
The source had fewer rankable highlights; the guidance is to return what came back with a note rather than padding with low-score filler.
Quick Setup:
.claude/skills/Repository
samuraigpt/generative-media-skills