HubLensAI AgentsArcReel/ArcReel
ArcReel

ArcReel

AI#AI Agent#Video Generation#FastAPI#React#Generative AI
View on GitHub
2,045

// summary

ArcReel is an open-source AI video generation workbench that implements an automated pipeline from novel scripts to finished videos via a multi-agent architecture. The platform supports integration with various providers including Gemini, Volcengine Ark, Grok, and OpenAI, offering character consistency maintenance and narrative tracking features. Users can manage projects, track costs, and export Jianying drafts through a visual interface to achieve efficient AI-assisted video creation.

// technical analysis

ArcReel is an open-source video generation workbench powered by AI Agents, designed to automatically convert text content such as novels into short videos. The project adopts a multi-agent architecture based on the Claude Agent SDK, achieving full-process automation from script creation and character design to video synthesis through a collaborative mode of orchestrated Skills and focused Subagents. Its core technical decision lies in introducing a unified backend abstraction layer that supports multiple providers such as Gemini, Volcano Engine Ark, Grok, and OpenAI, combined with an asynchronous task queue and a visual workbench to effectively solve common challenges in AI video generation such as consistency, task scheduling, and cost control.

// key highlights

01
Adopts a multi-agent architecture to automatically complete the entire production pipeline from script to video through the collaboration of orchestrated Skills and Subagents.
02
Supports multi-provider access, allowing flexible switching between mainstream AI models like Gemini, Volcano Engine Ark, Grok, and OpenAI at the global or project level.
03
Built-in character consistency and clue tracking features, ensuring visual coherence across shots through pre-generated character design sheets.
04
Provides an asynchronous task queue and a visual management interface, supporting breakpoint resumption, version rollback, and real-time task status tracking.
05
Equipped with detailed cost tracking and estimation functions, supporting the statistics of multimodal generation costs based on provider strategies.
06
Supports exporting to Jianying draft format, making it convenient for users to perform secondary creation and refined editing in professional video editing software.

// use cases

01
Multi-agent automated video generation workflow based on the Claude Agent SDK
02
Support for multi-provider image and video generation with character consistency and narrative tracking capabilities
03
Built-in visual workbench supporting project management, cost tracking, and one-click export of Jianying drafts

// getting started

Developers need to ensure the environment supports POSIX (such as Linux, MacOS, or Windows WSL2), clone the repository, enter the deploy directory, configure the .env file, and use docker compose up -d to start the service. After the first access, please configure the Anthropic API Key on the settings page to power the AI assistant, and add the API Keys for image or video generation providers according to your needs to start creating.