This project offers an AI-powered academic writing assistant for professional paper polishing and language optimization, also serving as an AIGC de-duplication tool to bypass AI detection services like GPTZero. It targets researchers and students aiming to enhance academic work quality and originality, ensuring text appears human-generated.

How It Works

It employs a two-stage optimization: paper polishing followed by academic enhancement for originality. The system recommends Google Gemini 2.5 Pro for its performance and cost-effectiveness across tasks like polishing, enhancement, and session compression. An OpenAI-compatible API format is used. The architecture features a FastAPI backend and React frontend, using SQLite/PostgreSQL and Redis for data and concurrency management, respectively.

Quick Start & Requirements

Installation uses unified scripts (start-macos.sh, start.ps1, start.sh) or manual command-line setup. Prerequisites include Python, Node.js, a database (SQLite/PostgreSQL), Redis, and API keys/base URLs for OpenAI/Gemini. Configuration is via a .env file. Access points: Frontend (http://localhost:3000), API Docs (http://localhost:8000/docs), Admin Panel (http://localhost:3000/admin).

Highlighted Details

• Dual-Stage Optimization: Separate polishing and academic enhancement phases.
• Intelligent Segmentation: Auto-detects titles and skips short paragraphs (configurable threshold).
• Usage Control: Implements a card (卡密) system with configurable usage limits.
• Concurrency Management: Utilizes queue management and dynamic adjustment of concurrent users.
• Admin Panel: Offers user stats, session monitoring, key generation, database management, and system configuration.

Maintenance & Community

The provided README lacks specific details on maintainers, community channels, or project roadmaps.

Licensing & Compatibility

Licensed under Creative Commons (CC BY-NC-SA 4.0). Commercial use is strictly prohibited without explicit permission ("未经允许禁止商业使用"), posing a significant adoption barrier for businesses.

Limitations & Caveats

Production deployment requires careful setup: changing default credentials, generating strong security keys, and ensuring valid API endpoints. The non-commercial license is a primary blocker for business adoption. Setup involves managing Python, Node.js, Redis, and database configurations, plus external API dependencies.