PAICNI/CFBox

一个运行在 Cloudflare Workers 上兼容CFnew 与edgetunnel的代理订阅管理面板——既能在 /UUID 路径上输出 VLESS/Trojan/xhttp 多协议订阅，又提供图形化配置界面（KV 存储、改完即生效），并内置延迟测试、流媒体/AI 连通性检测、多客户端订阅生成、多语言等能力。

What it solves

CFBox is a subscription manager for proxy services running on Cloudflare Workers or Pages. It simplifies the process of creating and managing VLESS, Trojan, and xhttp subscriptions by aggregating multiple IP sources and providing a user-friendly graphical interface for configuration, eliminating the need for manual environment variable editing.

How it works

The project deploys as a single JavaScript file on Cloudflare's edge network. It uses Cloudflare KV (Key-Value storage) to store configurations, allowing changes made in the web panel to take effect immediately. It merges IP sources from both CFnew and edgetunnel, combining official Cloudflare CIDR ranges, curated domain lists, and custom IP lists to generate a large number of optimized nodes. It automatically detects the user's client (e.g., Clash, Surge, Sing-box) via the User-Agent to deliver the correct subscription format.

Who it’s for

Users who want to set up their own proxy subscription service on Cloudflare Workers/Pages and need a way to easily manage nodes, optimize connection speeds via "preferred IP" (优选 IP) selection, and configure protocol settings without deep technical knowledge of the platform's internal variables.

Highlights

  • Dual-Mode Management Panel: Offers a "Standard Mode" for common tasks and an "Advanced Mode" with a neon-tech aesthetic for granular control.
  • Multi-Protocol Support: One-click toggles for VLESS, Trojan, and xhttp protocols.
  • Optimized Subscription Generation: Built-in support for 26 preferred IP sources, random official CIDR generation, and custom IP aggregation.
  • Broad Client Compatibility: Automatic recognition and subscription generation for 10+ popular clients including Clash, Shadowrocket, and Sing-box.
  • Integrated Network Testing: Built-in tools to test connectivity to AI services (GPT, Gemini) and streaming platforms (Netflix, Disney+), as well as a Google speed test.
  • Edge-Native Performance: Implements KV read optimization with 30-second memory caching to reduce latency and API pressure.

Related

  • Project
    unstablebuild/rune

    A fast, GPU-rendered, keyboard-driven IDE that integrates a modular AI coding agent and Unix-style composable tools for power users.

  • Project
    pygfx/wgpu-py

    A Python implementation of the WebGPU API that provides a powerful and reliable interface for GPU graphics and compute applications.

  • Project
    QwenLM/FlashQLA

    A high-performance linear attention kernel library that accelerates Gated Delta Rule (GDN) Chunked Prefill, providing significant speedups for training and inference on NVIDIA Hopper and Blackwell GPUs.

  • Project
    pegainfer-project/pegainfer

    PegaInfer is a Rust‑based, CUDA‑accelerated LLM inference server that provides an OpenAI‑compatible API, supports multiple model families (Qwen3, Gemma 4, DeepSeek‑V2‑Lite, etc.), and offers production‑grade performance without any Python runtime.