A desktop application that unifies access to various AI subscription services (like ChatGPT, Claude, Gemini, GitHub Copilot) for integration into any coding tool. It provides a user-friendly interface for managing AI connections, monitoring usage, and routing requests through a local proxy, benefiting developers seeking a consistent AI experience across different coding environments.

How It Works

ProxyPal leverages a Rust backend with Tauri v2 for its native desktop application framework, paired with a SolidJS frontend for the user interface. It bundles and utilizes CLIProxyAPI to manage connections to multiple AI providers, including OpenAI-compatible endpoints, Claude, Gemini, and GitHub Copilot. This architecture allows it to present a unified OpenAI-compatible API endpoint (http://localhost:8317/v1) that coding tools can connect to, abstracting away the complexities of individual AI service integrations and authentication.

Quick Start & Requirements

Download the application from the releases page. Launch ProxyPal, connect your AI accounts (via OAuth or auth files), and configure your coding tool to use http://localhost:8317/v1. For macOS users experiencing security prompts, run xattr -cr /Applications/ProxyPal.app in the terminal. Supported platforms include macOS (Apple Silicon and Intel), Windows (x64), and Linux (x64, .deb). Development requires pnpm and Rust.

Highlighted Details

• Supports numerous AI providers: Claude, ChatGPT, Gemini, Qwen, iFlow, Vertex AI, and custom OpenAI-compatible endpoints.
• Includes a GitHub Copilot bridge, enabling Copilot models to be accessed via an OpenAI-compatible API.
• Provides usage analytics to track requests, tokens, success rates, and estimated costs.
• Features auto-configuration to detect and set up installed CLI agents.

Maintenance & Community

The project appears to be maintained by an individual developer, indicated by the "buy me a coffee" link for support. No specific community channels (like Discord or Slack) or formal roadmap are detailed in the README.

Licensing & Compatibility

The project is licensed under the MIT license, which permits commercial use, modification, and distribution with minimal restrictions, requiring only attribution.

Limitations & Caveats

The macOS application is not yet signed with an Apple Developer certificate, requiring users to bypass security warnings manually. The README does not specify performance benchmarks or known bugs.