project
html-video - An open-source HTML version of CapCut from the Open Design team.
html-video is an open-source HTML version of CapCut, developed by the OpenDesign team (nexu.io) and built on the hyperframes framework. The tool allows agents to generate world-class product promotions and explanations by writing HTML...
What is html-video?
html-video is an open-source HTML version of CapCut, built on the hyperframes framework, developed by the Open Design team (nexu.io). The tool allows agents to generate world-class product promotions, educational videos, and more by writing HTML. It includes 20+ top-tier templates, supports paginated editing, multi-size export, and Minimax AI voiceover and background music, and can run locally without an API key.
The main functions of html-video
- Templated video generationIt has more than 20 built-in top-level HTML video style templates, covering scenarios such as product promotion, knowledge explanation, and data visualization. The Agent can directly call them to generate dynamic MP4s.
- Pagination editing and previewThe Studio editor supports paginated preview, paginated editing, and frame-level text modification, eliminating the need for repeated exporting to view the effects and significantly improving editing efficiency.
- Multi-size adaptationOne-click switching between landscape 16:9, portrait 9:16, square 1:1, Xiaohongshu 4:5 and other mainstream video formats to meet the distribution needs of all platforms.
- Seamless Agent IntegrationAutomatically recognizes 6 local Code Agent CLIs (Codex, Claude Code, Hermes, Cursor Agent, etc.), allowing one-click switching in the top bar, and is ready to use with zero configuration.
- AI voiceover and background musicIt has been integrated with Minimax and supports automatically generating narration and background music based on video content. Token consumption is low and costs are controllable.
How to use html-video
- Cloning repositoryPull the html-video open-source project from GitHub to your local machine.
- Install dependenciesRun the installation command and configure the local environment; no API key application is required.
- Launch StudioOpen the local visual editor, browse 20+ built-in templates, and choose the style that suits you.
- Input contentThe Agent will automatically generate an HTML video script by providing website links, files, or articles from public accounts.
- Pagination RefinementPreview and edit text and visuals page by page in Studio, and adjust details in real time.
- Export MP4After confirming everything is correct, you can export the high-definition video with one click, or directly generate the dubbing and background music and then download it.
The core advantages of html-video
- Local-first, zero API costIt runs entirely locally, does not rely on cloud API keys, ensures data privacy, and has extremely low long-term usage costs.
- Agent native designDesigned specifically for Code Agent workflows, it supports one-click switching of mainstream CLI tools, enabling AI to automatically complete the entire process from content to video.
- HTML-driven, extremely low costUsing HTML/CSS to replace traditional video engineering, a CapCut-level experience can be achieved with just 30,000 lines of code, with a technical threshold and maintenance cost far lower than professional video software.
- Real-time iteration, what you see is what you get.The paginated editing mode supports frame-by-frame modification and real-time preview, avoiding the inefficient cycle of "modification-rendering-inspection" in traditional video tools.
The project address for html-video
- GitHub repository: https://github.com/nexu-io/html-video
Comparison of similar products to html-video
| Dimension | html-video | FireRed-OpenStoryline |
|---|---|---|
| Core positioning | HTML version of CapCut, Agent native video generation | A conversational AI video editing agent with directorial thinking |
| Technical Architecture | hyperframes + HTML/CSS rendering engine | MoviePy + FFmpeg + LangChain + MCP protocol |
| Interaction methods | Studio visual editor + Agent CLI code-driven | Natural language dialogue + real-time interaction via web interface |
| Source of materials | User input (link, file, article) + 20+ built-in templates | Automatic online search (Pexels) + User-uploaded materials |
| Editing granularity | Pagination preview + frame-level text editing | Dialogue-based retouching (trimming, replacing, reorganizing, color correction) |
| AI capabilities | Template-based generation, Minimax voice-over and background music | Intelligent script generation, ASR rough cut, AI transitions, style imitation |
| Agent integration | It natively supports 6 CLIs (Codex/Claude Code/Cursor, etc.) | OpenClaw / Claude Code / Codex Skills Integration |
| Local/Cloud | Local-first approach, core functionality requires zero API key. | For local deployment (Docker/Conda), an LLM API Key needs to be configured. |
| Hardware Requirements | It can run on a regular laptop | The core logic can run without a GPU; the CPU is sufficient. |
| Templates/Skills | 20+ Built-in HTML Video Templates | Editing skills refinement (reuse of rhythm, color tone, and transition styles) |
| Output format | MP4 in multiple aspect ratios (16:9 / 9:16 / 1:1 / 4:5) | Custom resolution video |
| Open source license | Apache 2.0 | Apache 2.0 |
| Target audience | Developers, Agent workflows, product promotion/knowledge explanation | Content creators, Vlog bloggers, and producers of narrated videos |
Application scenarios of html-video
- Mass production of knowledge blogger contentAutomatically converts WeChat official account articles and technical documents into knowledge explanation videos with voiceovers, adapting to multiple platform specifications such as Xiaohongshu 4:5 and Bilibili 16:9.
- Product promotional videoEnter the product's official website link, and the Agent will automatically extract the selling points and generate a dynamic product demonstration video, making it suitable for startups to quickly create marketing materials.
- Data visualization reportTransform CSV/JSON data into dynamic chart videos for presentations and social media sharing, avoiding the boredom of static charts.
- E-commerce content creationBatch generate short videos such as product introductions and usage tutorials, and use Minimax AI voiceovers to achieve low-cost, high-efficiency content matrix operation.