What is video-remix-workflow?

macong0420/video-remix-workflow — explained in plain English

Analysis updated 2026-05-18

★ 16PythonAudience · vibe coderComplexity · 3/5Setup · moderate

Mindmap

mindmap
  root((video-remix-workflow))
    What it does
      Analyze source video
      Plan remix angle
      Draft voiceover script
      Render with ffmpeg
    Tech stack
      Python
      ffmpeg
      Codex skill
    Use cases
      Create commentary videos
      Track source citations
      Build storyboards
      Automate ffmpeg rendering
    Audience
      Video creators
      Codex users
    Safety limits
      Max 60 percent footage reuse
      No hidden sources
      No voice cloning
    Getting started
      Clone into skills dir
      Confirm render plan

mindmap root((video-remix-workflow)) What it does Analyze source video Plan remix angle Draft voiceover script Render with ffmpeg Tech stack Python ffmpeg Codex skill Use cases Create commentary videos Track source citations Build storyboards Automate ffmpeg rendering Audience Video creators Codex users Safety limits Max 60 percent footage reuse No hidden sources No voice cloning Getting started Clone into skills dir Confirm render plan

What do people build with it?

USE CASE 1

Turn an existing video into a short commentary remix with a planned angle and script.

USE CASE 2

Generate a storyboard and asset checklist before rendering a remix.

USE CASE 3

Automatically render a confirmed plan into a final video using ffmpeg.

USE CASE 4

Keep a record of source citations, licensing notes, and footage usage limits.

What is it built with?

PythonffmpegOpenAI Codex

How does it compare?

	macong0420/video-remix-workflow	920linjerry-stack/capital-studio	adya84/ha-world-cup-2026
Stars	16	16	16
Language	Python	Python	Python
Setup difficulty	moderate	easy	easy
Complexity	3/5	3/5	2/5
Audience	vibe coder	researcher	general

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Rendering needs Python 3.10+, ffmpeg, and ffprobe installed, plus a confirmed render plan file.

No license has been declared for this project yet, so no usage rights are granted beyond viewing the source.

So what is it?

video-remix-workflow is a skill for OpenAI Codex that guides a creator through producing a remixed video from an existing source. A skill in this context is a plugin that teaches an AI coding assistant how to handle a specific kind of task. When installed, you can ask Codex to take an original video and help you turn it into a new piece of content, such as a 60-second commentary. The workflow covers several stages. It starts by analyzing the source material and recording where it came from, including the original title, URL, author, and license notes. It then helps plan the remix from a chosen angle, drafts a voiceover script, builds a storyboard, and assembles a list of clips and other assets to use. Once those pieces are confirmed by the user, the workflow generates a structured plan file and passes it to a Python script that uses ffmpeg (a standard video processing tool) to assemble the final video. The tool has a built-in safety model. It limits how much of the original video's footage can appear in the remix, with a default cap of 60 percent. It requires source citations for any clips used, and blocks requests to hide sources, remove watermarks, or disguise borrowed material as original work. Source audio from the original is off by default, and voice cloning of the original host is not supported without explicit authorization. Installation involves cloning the repository into the Codex skills directory. The rendering step requires Python 3.10 or newer and ffmpeg. The README notes that no software license has been declared for this project yet.

Copy-paste prompts

Prompt 1

Explain how video-remix-workflow's safety model limits how much original footage can be reused.

Prompt 2

Help me install video-remix-workflow into my Codex skills directory.

Prompt 3

Walk me through building a render_plan.json for a 60-second commentary remix.

Prompt 4

Show me how to run render_remix.py in dry-run mode to validate a plan before rendering.

Frequently asked questions