AutoShorts is an AI-powered package for generating short-form videos (YouTube Shorts, TikTok) by automating scriptwriting, voiceovers, and image selection. It targets developers and content creators seeking to streamline AI-driven video production.

How It Works

It employs a flexible abstraction layer integrating multiple AI tools for script generation (Ollama, OpenAI, Gemini, Claude), TTS (ElevenLabs, local), and image sourcing (Pexels, scraper). This modular design facilitates extension with new AI models and video types. AutoShorts offers a CLI, a JS API for programmatic use, and an experimental web UI.

Quick Start & Requirements

Install the CLI globally via npm install -g auto-shorts. For the JS interface/UI, clone the repo and run npm install, npm run install-ui-deps. Prerequisites include Node.js, npm, and optionally Ollama (ollama pull llama3.2) for local LLMs. API keys for cloud services (OpenAI, ElevenLabs, Pexels) are necessary for non-local AI. Docker installation is experimental. Setup involves cloning, dependency installation, and .env configuration.

Highlighted Details

• Supports diverse video formats: TTS, topic-based (news, educational), text message stories, 'Would you rather', ranking items.
• Integrates multiple AI services for scripts, voice, and images.
• Provides CLI, JS API, and an experimental web UI.
• Enables local LLM usage via Ollama for script generation.

Maintenance & Community

Active development has ceased; the project is maintained only for minor bug fixes. Contributions for new video types or AI tools are welcomed via GitHub issues/PRs, but no dedicated community channels (Discord/Slack) are listed.

Licensing & Compatibility

The project's license is not explicitly stated, hindering compatibility assessment, particularly for commercial use. It is marked "Not meant for production usage" with a "Use at your own risk" disclaimer and potential for breaking changes.

Limitations & Caveats

AutoShorts is in an early development stage with potential bugs, especially regarding unpredictable AI output requiring manual JSON correction. Active development has stopped, limiting future enhancements. The absence of a specified license is a significant adoption blocker.