davidcervinka/vibe-editing — explained in plain English
Analysis updated 2026-05-18
Create a product launch film by dropping raw clips into a folder and telling Claude what pacing and music you want.
Produce social media reels with captions and voiceover by describing the style in plain English.
Build an event aftermovie by having Claude inventory your footage, propose an edit plan, and refine it through conversational notes.
Generate a feature walkthrough video with split-screens, motion graphics, and background music using AI-assisted editing.
| davidcervinka/vibe-editing | 1ncendium/aibuster | aaronmayeux/ha-hurricane-tracker | |
|---|---|---|---|
| Stars | 5 | 5 | 5 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | moderate | easy |
| Complexity | 3/5 | 3/5 | 2/5 |
| Audience | vibe coder | ops devops | general |
Figures from each repo's GitHub metadata at analysis time.
Requires installing two external open-source projects first (video-use and HyperFrames) plus API keys for ElevenLabs and RunwayML, and you must supply your own raw video clips.
{"i":0,"repo":"davidcervinka/vibe-editing","stars":5,"lang":"Python","ok":true,"char_count":1879,"text":"Vibe-editing is a method for producing videos by talking to an AI assistant instead of using a traditional video editor. You describe what you want in plain English: the cuts, the pacing, the music, the captions, the motion. Claude Code then coordinates a collection of tools to build the video for you. It can produce product launch films, event aftermovies, feature walkthroughs, and social media reels.\n\nThe project coordinates several tools across different roles. Claude Code acts as the orchestrator, reading your intent, planning the edit, running commands, and reviewing its own output. ffmpeg handles the actual media work: cutting, scaling, compositing, and combining clips. ElevenLabs provides original background music, voiceover narration, and word-level transcription for captions. Python with Pillow generates graphics like end cards and caption overlays. HyperFrames or Remotion handle motion graphics.\n\nThe workflow follows a loop. You drop raw clips into a folder, Claude inventories them and proposes a plan in plain English. After you confirm, it extracts and normalizes each segment to a uniform format, composes the video with overlays and split-screens, adds audio, and generates captions last. Then it samples frames from its own output and checks audio loudness before showing you the result. You give notes, and the cycle repeats until you are satisfied.\nThe repository includes shell and Python scripts for each stage of this pipeline. There are scripts for normalizing footage, building split-screens, generating music and voiceover, transcribing audio, creating captions, animating end cards, color grading, and mixing audio. It also includes AI b-roll generation through the RunwayML API. The repo ships scripts and documentation only, not source footage or brand assets. You bring your own clips.\n\nSetup requires installing two separate open-source projects first: video-use, which provides the conversational editing skill for Claude Code, and HyperFrames, which handles motion graphics. The README points to external repositories for both. Additional documentation files cover the tool stack, workflow details, editing techniques, motion design, generative footage, common pitfalls, and a reference library of real launch videos to study.\n\nThe full README is longer than what was shown.","error":null}
Edit videos by chatting with an AI assistant in plain English. Describe cuts, pacing, music, and captions, and Claude Code coordinates ffmpeg, ElevenLabs, and Python tools to build the video for you.
Mainly Python. The stack also includes Python, ffmpeg, ElevenLabs.
No license information is provided in the repository, so default copyright restrictions may apply.
Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
double-check against the repo, no cap.