liuluhaixiu/davinci-autoedit-agent — explained in plain English
Analysis updated 2026-05-18
Turn a folder of raw video, audio, and image clips into a drafted edit blueprint and timeline automatically.
Generate narration and titles grounded strictly in what the footage actually shows.
Get a list of missing shots still needed after comparing the assembled cut against the original script.
| liuluhaixiu/davinci-autoedit-agent | anthonykhayesaudsrx50512/flash-usdt-sender | confluentinc/quickstart-streaming-agents | |
|---|---|---|---|
| Stars | 79 | 78 | 78 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | moderate | hard |
| Complexity | 3/5 | 3/5 | 4/5 |
| Audience | vibe coder | general | developer |
Figures from each repo's GitHub metadata at analysis time.
Needs Python 3.10+ and FFmpeg, DaVinci Resolve Studio and AI/TTS API keys are optional.
DaVinci AutoEdit Agent is a Python-based tool that automates video editing inside DaVinci Resolve, a professional video editing application made by Blackmagic Design. It installs as a set of skills for Codex, an AI coding assistant, and guides you from raw media files all the way to a finished cut. The workflow runs in stages, and each stage requires your explicit approval before the next one starts. You give the agent a folder of video, audio, or image files and describe what you want to make. It scans the media, reviews frames, drafts a script, optionally generates a voice-over, builds an edit blueprint, then creates a full timeline inside DaVinci Resolve. After the cut is assembled, it compares the result against the original script and lists any missing shots that still need to be filmed. Three skills are bundled together. The main skill handles the complete editing pipeline. A second one writes narration and titles grounded only in what the footage actually contains. The third controls DaVinci Resolve directly, verifying that each write succeeds and that the original source files are never modified. To run the tool you need Python 3.10 or newer and FFmpeg installed on your machine. DaVinci Resolve Studio 18.5 or newer is optional. Without it, the agent can still analyze footage, write a script, and produce an edit blueprint you carry into your editor manually. API keys for an AI model and a text-to-speech service are also optional. If you skip text-to-speech, the agent designs the cut around production audio, music, captions, or silence instead. The project is released under the MIT license and is not affiliated with Blackmagic Design.
A Codex skill that automates editing raw video footage into a finished cut inside DaVinci Resolve.
Mainly Python. The stack also includes Python, FFmpeg, DaVinci Resolve.
MIT license: use, copy, modify, and distribute freely, including commercially, as long as you keep the copyright notice.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.