DocuScribe

Docuscribe is a Next.js application designed for scraping, managing, and organizing web documentation. It leverages AI-powered flows for tasks like hashtag generation and content parsing.

I also wrote an MCP server for it to release its full power in vibe coding: see docuscribe-mcp-server

Features

• Web Scraping: Extracts documentation from websites, following links to compile structured content.
• AI Integration: Generates hashtags and summaries using AI providers.
• Document Management: Allows users to rename, delete, schedule updates, and export documentation.
• Scheduling: Supports daily, weekly, and monthly update schedules for automated scraping.
• Library View: Browse and search scraped documentation with filters and detailed views.
• Customizable UI Components: Includes reusable components like buttons, dialogs, and schedule selectors.

Project Structure

• src/app/: Contains Next.js pages and API routes.
  • page.tsx: Main scraper interface.
  • library/page.tsx: Documentation library view.
  • schedule/page.tsx: Scheduled tasks management.
• src/components/: Reusable UI components.
  • ui/: Includes buttons, dialogs, inputs, and more.
  • schedule/: Contains the ScheduleSelect component.
• src/lib/: Shared utilities and types.
  • db.ts: Database entity definitions.
  • content.ts: Markdown section parsing helpers.
  • utils.ts: Utility functions like class merging and numeric validation.
• src/state/: Global state management for scraping jobs.
• scripts/: Test scripts for scraping logic.

Getting Started

1. Install dependencies:

   npm install

2. Run the development server:

   npm run dev

3. Build for production:

   npm run build

Screenshots

Main Panel

Library

Scheduled Update

Notes

• The application uses SQLite for local data storage.
• AI flows are powered by Genkit.
• TailwindCSS is used for styling.

For more details, explore the codebase starting with src/app/page.tsx. Contributions and feedback are welcome!