DocGenius is a smart, 4-step AI-powered document generation tool. From a simple topic idea, it helps you generate structured outlines, detailed chapter plans, and polished full-text drafts — all with AI assistance at every stage.
-
🧠 AI-Powered Outlining — Enter a topic and optional writing style; AI generates 3 distinct outlines with chapter structures. Supports diverse tones (Academic, Professional, Creative, etc.).
-
📝 Smart Chapter Breakdown — Each chapter gets a detailed bullet-point plan generated by AI, covering sub-topics, arguments, and data points to include.
-
✍️ One-Click Chapter Drafting — AI writes full, coherent content for every chapter based on the detailed plan. Supports Markdown formatting and optional chart/image analysis for data-driven sections.
-
🔧 Multi-Provider Support — Defaults to Google Gemini (2.5 Flash / Pro) with built-in support for any OpenAI-compatible API (Moonshot, DeepSeek, custom endpoints, etc.).
-
📤 Export & Refine — Review the final document in-app, refine content with custom rewrite instructions, and export your finished work.
-
📱 PWA Ready — Installable as a standalone app with offline support.
| Layer | Technology |
|---|---|
| Framework | React 19 + TypeScript |
| Build Tool | Vite 6 |
| Styling | Tailwind CSS (via CDN) |
| AI SDK | @google/genai (Google Gemini) |
| Alternate AI | OpenAI-compatible REST API |
| PWA | Service Worker + Web Manifest |
TextGen/
├── App.tsx # Root component & global state
├── index.tsx # Entry point
├── index.html # HTML shell with Tailwind CDN
├── types.ts # TypeScript type definitions & enums
├── vite.config.ts # Vite configuration
├── manifest.json # PWA manifest
├── sw.js # Service Worker
├── components/
│ ├── Step1_Topic.tsx # Step 1: Topic input & outline selection
│ ├── Step2_Details.tsx # Step 2: Chapter detail planning
│ ├── Step3_Drafting.tsx # Step 3: AI content generation
│ ├── Step4_Export.tsx # Step 4: Review, refine & export
│ ├── StepIndicator.tsx # Progress step indicator
│ └── SettingsModal.tsx # AI provider & model configuration
└── services/
└── geminiService.ts # AI call abstraction (Gemini SDK + OpenAI fetch)
- Node.js >= 18
- A Gemini API Key from Google AI Studio (or an OpenAI-compatible endpoint)
# Clone the repository
git clone https://github.com/pengcong2020520/TextGen.git
cd TextGen
# Install dependencies
npm installCreate a .env.local file in the project root:
GEMINI_API_KEY=your_gemini_api_key_hereYou can also switch to a custom AI provider at runtime via the Settings panel in the app — no additional configuration required.
npm run devOpen http://localhost:3000 in your browser.
npm run build
npm run preview # Preview the production buildStep 1 Step 2 Step 3 Step 4
┌──────────────┐ ┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ Topic & │ → │ Chapter │ → │ AI │ → │ Review & │
│ Outline │ │ Details │ │ Drafting │ │ Export │
│ │ │ │ │ │ │ │
│ • Enter topic│ │ • AI fills │ │ • AI writes │ │ • Refine │
│ • Pick style │ │ bullet │ │ full text │ │ • Copy/export│
│ • AI generates│ │ points per │ │ per chapter│ │ • Start over │
│ 3 outlines │ │ chapter │ │ • Chart │ │ │
│ • Edit as │ │ • Edit as │ │ analysis │ │ │
│ needed │ │ needed │ │ support │ │ │
└──────────────┘ └──────────────┘ └──────────────┘ └──────────────┘
The app is deployed on Vercel at text-gen-theta.vercel.app. To deploy your own instance:
This project is open source and available under the MIT License.
Made with ❤️ by @pengcong2020520