Pandora Helper is a self-hosted web application designed to facilitate the sharing of ChatGPT Plus and Claude Pro services among users. It provides centralized account management, granular sharing controls, and a user-friendly interface that mirrors the official ChatGPT and Claude web UIs, enabling efficient and controlled access to premium AI models for teams or groups.

How It Works

The project leverages existing open-source projects (oaifree and fuclaude) as its foundation. It manages user accounts by storing refresh tokens and access tokens, allowing for automatic token refreshing to maintain service continuity. The core functionality revolves around creating shareable tokens with customizable limits on usage, session duration, and model access, ensuring fair distribution and preventing abuse.

Quick Start & Requirements

• Docker: docker run -d --name PandoraHelper --net=bridge -p 9000:9000 -v ./data:/app/data q11391/pandora-helper
• Prerequisites: Docker, or manual download for Linux/Unix/Windows. A config.json file is required for configuration, notably setting an admin_password.
• Setup: Minimal setup time, primarily involving configuration.
• Documentation: https://wiki.oai2b.com

Highlighted Details

• Mirrors official ChatGPT/Claude web UI for a familiar user experience.
• Supports session isolation, usage quotas, expiry limits, and ethical restrictions for shared accounts.
• Features usage statistics tracking for shared accounts.
• Automatic token refreshing mechanism for sustained access.

Maintenance & Community

The project acknowledges contributions from LinuxDo and uses Slash-Admin for its frontend and go-nunu for its backend. Links to community or roadmap are not explicitly provided in the README.

Licensing & Compatibility

The project's licensing is not explicitly stated in the README. Compatibility for commercial use or closed-source linking is not detailed.

Limitations & Caveats

The README does not specify licensing details, which may impact commercial adoption. While it supports multiple deployment methods (manual, Docker, Kubernetes), detailed troubleshooting or advanced configuration beyond the provided config.json is not elaborated upon.