OpusClip
OpusClipLong video to short clips, with captions and reframing.
Align subtitles to speech without drifting.
Captioned video where a transcript exists but no timecodes
Translation quality — timing and wording are separate passes
Takes a transcript and produces cue-level timecodes: short lines, under two seconds where possible, no more than two lines on screen. The check is mechanical — every cue starts before its audio and ends after it — which is what separates a usable skill from a guess with timestamps.
Clone the repository, then copy the skill folder into your agent's skills directory — ~/.claude/skills/ for Claude Code. Skills are plain files: no build step, no daemon.
npx skills add subtitle-timingRead the SKILL.md before you install. A good skill states when it applies, what it must not do, and how to verify the result. A skill that only lists adjectives will change the agent's tone without improving its output — skip those.
Long video to short clips, with captions and reframing.
OpenAI's video model: text or image in, short clip out, with storyboard control.
Google DeepMind's video model, available through Gemini and Vertex AI.
A server from Mux's SDK monorepo that exposes video assets, live streams, and playback IDs. The read path is what earns its place: an agent can answer why a video is…
Wraps the YouTube Data API for search, channel and video metadata, and captions where available. The transcript path is the useful one for research; anything that re…
Runs Deque's axe engine against a page and returns structured violations with the failing selector and rule. It will not find everything — automated checks catch rou…