69.2K ·GitHub
What is this?
OpenMontage turns your AI coding assistant (Claude Code, Cursor, Copilot, Windsurf, Codex) into a full video production studio: you describe a video in plain language and it handles research, scripting, asset generation, editing, and final composition through 12 production pipelines and 100+ tools.
OpenMontage turns your AI coding assistant (Claude Code, Cursor, Copilot, Windsurf, Codex) into a full video production studio: you describe a video in plain language and it handles research, scripting, asset generation, editing, and final composition through 12 production pipelines and 100+ tools.Is this an app I run, or a toolkit my coding agent uses?It is not a standalone app with a UI you click through — it is a Python toolkit of pipelines, skills, and production-knowledge files that your existing AI coding assistant reads and operates, plus a local 'Backlot' web dashboard that visualizes the pipeline as it runs.
It is not a standalone app with a UI you click through — it is a Python toolkit of pipelines, skills, and production-knowledge files that your existing AI coding assistant reads and operates, plus a local 'Backlot' web dashboard that visualizes the pipeline as it runs.How do I get it running?Clone the repo, run the provided setup (or manually: create a Python venv, pip install requirements, npm install in remotion-composer, pip install piper-tts, copy .env.example to .env), then open the project in your AI coding assistant and just tell it what video you want.
Clone the repo, run the provided setup (or manually: create a Python venv, pip install requirements, npm install in remotion-composer, pip install piper-tts, copy .env.example to .env), then open the project in your AI coding assistant and just tell it what video you want.Do I need an API key for any of this?No — out of the box it works with zero paid API keys using Piper TTS for narration, Archive.org/NASA/Wikimedia for free footage, and Pexels/Unsplash/Pixabay for stock; adding keys (e.g. for Veo, Kling, ElevenLabs, Flux, Stable Diffusion) just unlocks more tools and models.
No — out of the box it works with zero paid API keys using Piper TTS for narration, Archive.org/NASA/Wikimedia for free footage, and Pexels/Unsplash/Pixabay for stock; adding keys (e.g. for Veo, Kling, ElevenLabs, Flux, Stable Diffusion) just unlocks more tools and models.Does it actually use real video footage, or is everything animated still images?It supports both: by default it often builds image-based videos animated via Remotion, but it also has a free, real-footage path that pulls actual motion clips from Archive.org, NASA, and Wikimedia Commons — you need to explicitly ask for 'real footage only' to get that path.
It supports both: by default it often builds image-based videos animated via Remotion, but it also has a free, real-footage path that pulls actual motion clips from Archive.org, NASA, and Wikimedia Commons — you need to explicitly ask for 'real footage only' to get that path.Which coding assistants does it work with?Any AI coding assistant that can read files and run code — Claude Code, Cursor, Copilot, Windsurf, and Codex are explicitly named.
Any AI coding assistant that can read files and run code — Claude Code, Cursor, Copilot, Windsurf, and Codex are explicitly named.Does it run on Windows too, or just Mac/Linux?Yes, Windows is supported alongside macOS and Linux, with separate PowerShell setup instructions provided.
Yes, Windows is supported alongside macOS and Linux, with separate PowerShell setup instructions provided.Does any of this run locally without sending data out?Narration can run fully offline via Piper TTS, and with a GPU you can unlock free local video generation instead of calling paid cloud APIs.
Narration can run fully offline via Piper TTS, and with a GPU you can unlock free local video generation instead of calling paid cloud APIs.Can it start from a video I already like instead of a blank prompt?Yes — paste a YouTube video, Short, Reel, or TikTok link and the agent analyzes transcript, pacing, scenes, and style to generate 2-3 differentiated concepts with cost estimates before producing anything.
Yes — paste a YouTube video, Short, Reel, or TikTok link and the agent analyzes transcript, pacing, scenes, and style to generate 2-3 differentiated concepts with cost estimates before producing anything.What is Backlot?Backlot is a local 'living storyboard' dashboard that fills in automatically as a production runs, showing script pages, scene cards, asset generation, provider decisions, and costs, and lets you approve visuals mid-pipeline or replay a finished run.
Backlot is a local 'living storyboard' dashboard that fills in automatically as a production runs, showing script pages, scene cards, asset generation, provider decisions, and costs, and lets you approve visuals mid-pipeline or replay a finished run.What is Monty the Clapper?Monty the Clapper is the official mascot of OpenMontage, often appearing as a closing element in example productions.
Monty the Clapper is the official mascot of OpenMontage, often appearing as a closing element in example productions.How much does a typical video cost to produce?Costs vary by complexity and tools used — example productions in the README range from about $1.33 for a short animated piece to around $5 for a multi-world showcase, with free options available if using only open-source/free tools.
Costs vary by complexity and tools used — example productions in the README range from about $1.33 for a short animated piece to around $5 for a multi-world showcase, with free options available if using only open-source/free tools.What are similar projects?OpenReels, ai-promo-video-maker, storycast, brag, mortiflix-oss, AdCraft, WanGP, reelmimic
OpenReels, ai-promo-video-maker, storycast, brag, mortiflix-oss, AdCraft, WanGP, reelmimicWhich topics is it in?flux, remotion, video-production, stable-diffusion, video-editing, copilot, text-to-video, text-to-speech, ffmpeg, video-generation, image-generation, cursor
flux, remotion, video-production, stable-diffusion, video-editing, copilot, text-to-video, text-to-speech, ffmpeg, video-generation, image-generation, cursor