Insight Weaver

A serverless geopolitical analysis and publishing tool built on the Cloudflare stack.

Overview

Insight Weaver streamlines the process of geopolitical analysis by aggregating information from RSS feeds, enabling expert curation and editing, integrating AI for summarization, and generating polished reports.

Architecture

• Frontend: React with Vite and Tailwind CSS (deployed on Cloudflare Pages)
• Backend: Cloudflare Workers
• Database: Cloudflare D1 (SQL)
• Vector Search: Cloudflare Vectorize
• AI: Cloudflare Workers AI

Project Structure

insight-weaver/
├── frontend/          # React frontend application
│   ├── src/
│   │   ├── components/
│   │   │   ├── dashboard/     # The Ticker components
│   │   │   ├── workspace/     # The Canvas components  
│   │   │   ├── settings/      # Feed management UI
│   │   │   └── common/        # Shared components
│   │   ├── pages/            # Main pages
│   │   ├── hooks/            # React hooks
│   │   ├── utils/            # Utility functions
│   │   └── types/            # TypeScript types
│   └── public/               # Static assets
├── workers/           # Cloudflare Workers
│   ├── src/          # Worker source code
│   └── schema/       # Database schemas
└── docs/             # Documentation

Features

Core Workflow

1. Feed Management - Add and categorize RSS feeds
2. Content Ingestion - Dashboard view of incoming articles
3. Curation Workspace - Drag-and-drop interface for report building
4. AI Analysis - Summarization and analysis using Cloudflare Workers AI
5. Report Generation - Export to HTML, PDF, Markdown, and JSON formats

Key Components

• The Source Hub: RSS feed management and categorization
• The Ticker: Multi-column dashboard for incoming articles
• The Canvas: Drag-and-drop curation workspace
• The Analyst AI: AI-powered summarization and analysis
• The Publisher: Report generation and export functionality

Getting Started

1. Install dependencies:

   npm run setup

2. Start development server:

   npm run dev

3. Deploy to Cloudflare:

   npm run deploy

Configuration

The application requires Cloudflare account setup with:

• Cloudflare Pages (for frontend)
• Cloudflare Workers (for backend)
• Cloudflare D1 (for database)
• Cloudflare Vectorize (for semantic search)
• Cloudflare Workers AI (for summarization)

License

MIT License - see LICENSE file for details