by yeates · Agent Tool · ★ 59
Aurora Unified Video Editing with a Tool-Using Agent Aurora is an agentic video-editing framework. Given a source video and a natural language request, a tool-using vision-language agent first resolves ambiguity: it rewrites the request into an editor-facing instruction, decides the edit type, optionally retrieves a reference image from the web, and optionally grounds a target object with a mask. A unified video diffusion editor then performs the actual edit. The editor uses a WAN2.2-TI2V-5B diffusion transformer, a WAN2.2 VAE, and a frozen Qwen3.5-4B multimodal LLM.
| Stars | 59 |
| Forks | 2 |
| Language | Python |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 50.7285829893472/100 |
| Open Issues | 2 |
| Last Updated | 2026-06-16 |
| Created | 2026-05-18 |
| Platforms | python |
| Est. Tokens | ~13k |
These tools work well together with Aurora for enhanced workflows:
Explore other popular agent tool tools:
Aurora is Aurora: Unified Video Editing with a Tool-Using Agent. It is categorized as a Agent Tool with 59 GitHub stars.
Aurora is primarily written in Python.
You can find installation instructions and usage details in the Aurora GitHub repository at github.com/yeates/Aurora. The project has 59 stars and 2 forks, indicating an active community.
Aurora is released under the MIT license, making it free to use and modify according to the license terms.