docs: add bilingual README (Chinese + English) with non-commercial license

README.md

@@ -1,41 +1,54 @@
-# Codex Proxy
+<div align="center">
 
-A reverse proxy that exposes the [Codex Desktop](https://openai.com/codex) API as an OpenAI-compatible `/v1/chat/completions` endpoint. Use any OpenAI-compatible client (Cursor, Continue, VS Code, etc.) with Codex models — for free.
+  <h1>Codex Proxy</h1>
+  <h3>您的本地 Codex 编程助手中转站</h3>
+  <p>将 Codex Desktop 的能力以 OpenAI 标准协议对外暴露，无缝接入任意 AI 客户端。</p>
 
-## Architecture
+  <p>
+    <img src="https://img.shields.io/badge/Runtime-Node.js_18+-339933?style=flat-square&logo=nodedotjs&logoColor=white" alt="Node.js">
+    <img src="https://img.shields.io/badge/Language-TypeScript-3178C6?style=flat-square&logo=typescript&logoColor=white" alt="TypeScript">
+    <img src="https://img.shields.io/badge/Framework-Hono-E36002?style=flat-square" alt="Hono">
+    <img src="https://img.shields.io/badge/License-Non--Commercial-red?style=flat-square" alt="License">
+  </p>
 
-```
-OpenAI-compatible client
-        │
-   POST /v1/chat/completions
-        │
-        ▼
-  ┌─────────────┐     POST /backend-api/codex/responses
-  │ Codex Proxy │ ──────────────────────────────────────► chatgpt.com
-  │  :8080      │ ◄──────────────────────────────────────  (SSE stream)
-  └─────────────┘
-        │
-   SSE chat.completion.chunk
-        │
-        ▼
-    Client
-```
+  <p>
+    <a href="#-快速开始-quick-start">快速开始</a> •
+    <a href="#-核心功能-features">核心功能</a> •
+    <a href="#-技术架构-architecture">技术架构</a> •
+    <a href="#-客户端接入-client-setup">客户端接入</a> •
+    <a href="#-配置说明-configuration">配置说明</a>
+  </p>
+
+  <p>
+    <strong>简体中文</strong> |
+    <a href="./README_EN.md">English</a>
+  </p>
+
+</div>
+
+---
 
-The proxy translates OpenAI Chat Completions format to the Codex Responses API format, handles authentication (OAuth PKCE), multi-account rotation, and Cloudflare bypass via curl subprocess.
+**Codex Proxy** 是一个轻量级本地中转服务，将 [Codex Desktop](https://openai.com/codex) 的 Responses API 转换为 OpenAI 标准的 `/v1/chat/completions` 接口。通过本项目，您可以在 Cursor、Continue、VS Code 等任何兼容 OpenAI 协议的客户端中直接使用 Codex 编程模型。
 
-## Quick Start
+只需一个 ChatGPT 账号，配合本代理即可在本地搭建一个专属的 AI 编程助手网关。
+
+## 🚀 快速开始 (Quick Start)
 
 ```bash
-# 1. Install dependencies
+# 1. 克隆仓库
+git clone https://github.com/icebear0828/codex-proxy.git
+cd codex-proxy
+
+# 2. 安装依赖
 npm install
 
-# 2. Start the proxy (dev mode with hot reload)
+# 3. 启动代理（开发模式，支持热重载）
 npm run dev
 
-# 3. Open the dashboard and log in with your ChatGPT account
+# 4. 打开浏览器访问控制面板，使用 ChatGPT 账号登录
 #    http://localhost:8080
 
-# 4. Test a chat completion
+# 5. 测试请求
 curl http://localhost:8080/v1/chat/completions \
   -H "Content-Type: application/json" \
   -d '{
@@ -45,98 +58,90 @@ curl http://localhost:8080/v1/chat/completions \
   }'
 ```
 
-## Features
+## 🌟 核心功能 (Features)
 
-- **OpenAI-compatible API** — drop-in replacement for `/v1/chat/completions` and `/v1/models`
-- **OAuth PKCE login** — native browser-based login, no manual token copying
-- **Multi-account rotation** — add multiple ChatGPT accounts with automatic load balancing (`least_used` or `round_robin`)
-- **Auto token refresh** — JWT tokens are refreshed automatically before expiry
-- **Cloudflare bypass** — all upstream requests use curl subprocess with native TLS
-- **Quota monitoring** — real-time Codex usage/quota display per account
-- **Web dashboard** — manage accounts, view usage, and monitor status at `http://localhost:8080`
-- **Auto-update detection** — polls the Codex Desktop appcast for new versions
+### 1. 🔌 全协议兼容 (OpenAI-Compatible API)
+- 完全兼容 `/v1/chat/completions` 和 `/v1/models` 端点
+- 支持 SSE 流式输出，可直接对接所有 OpenAI SDK 和客户端
+- 自动完成 Chat Completions ↔ Codex Responses API 双向协议转换
 
-## Available Models
+### 2. 🔐 账号管理与智能轮换 (Auth & Multi-Account)
+- **OAuth PKCE 登录** — 浏览器一键授权，无需手动复制 Token
+- **多账号轮换** — 支持 `least_used`（最少使用优先）和 `round_robin`（轮询）两种调度策略
+- **Token 自动续期** — JWT 到期前自动刷新，无需人工干预
+- **配额实时监控** — 控制面板展示各账号剩余用量
 
-| Model ID | Alias | Description |
-|----------|-------|-------------|
-| `gpt-5.3-codex` | `codex` | Latest frontier agentic coding model (default) |
-| `gpt-5.2-codex` | — | Previous generation coding model |
-| `gpt-5.1-codex-max` | `codex-max` | Maximum capability coding model |
-| `gpt-5.2` | — | General-purpose model |
-| `gpt-5.1-codex-mini` | `codex-mini` | Lightweight, fast coding model |
+### 3. 🛡️ 反检测与协议伪装 (Anti-Detection)
+- **Chrome TLS 指纹** — 通过 curl-impersonate 模拟 Chrome 136 完整 TLS 握手特征
+- **桌面端请求头复现** — `originator`、`User-Agent`、`sec-ch-*` 等请求头按真实 Codex Desktop 顺序排列
+- **桌面上下文注入** — 每个请求自动注入 Codex Desktop 的系统提示词，实现完整的功能对等
+- **Cookie 持久化** — 自动捕获并回传 Cloudflare Cookie，维持会话连续性
+- **时间抖动 (Jitter)** — 定时操作加入随机偏移，消除机械化行为特征
 
-## API Usage
+### 4. 🔄 会话管理与版本追踪 (Session & Version)
+- **多轮对话关联** — 自动维护 `previous_response_id`，保持上下文连贯
+- **Appcast 版本追踪** — 定时轮询 Codex Desktop 更新源，自动同步 `app_version` 与 `build_number`
+- **Web 控制面板** — 账号管理、用量监控、状态总览，一站式操作
 
-### Chat Completions (streaming)
+## 🏗️ 技术架构 (Architecture)
 
-```bash
-curl http://localhost:8080/v1/chat/completions \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "codex",
-    "messages": [
-      {"role": "system", "content": "You are a helpful coding assistant."},
-      {"role": "user", "content": "Write a Python function to check if a number is prime."}
-    ],
-    "stream": true
-  }'
 ```
-
-### Chat Completions (non-streaming)
-
-```bash
-curl http://localhost:8080/v1/chat/completions \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "codex",
-    "messages": [{"role": "user", "content": "Hello!"}],
-    "stream": false
-  }'
-```
-
-### List Models
-
-```bash
-curl http://localhost:8080/v1/models
-```
-
-### Check Account Quota
-
-```bash
-curl "http://localhost:8080/auth/accounts?quota=true"
+                            Codex Proxy
+┌─────────────────────────────────────────────────────┐
+│                                                     │
+│  Client (Cursor / Continue / SDK)                   │
+│       │                                             │
+│  POST /v1/chat/completions                          │
+│       │                                             │
+│       ▼                                             │
+│  ┌──────────┐    ┌───────────────┐    ┌──────────┐  │
+│  │  Routes   │──▶│  Translation  │──▶│  Proxy   │  │
+│  │  (Hono)  │   │ OpenAI→Codex  │   │ curl TLS │  │
+│  └──────────┘   └───────────────┘   └────┬─────┘  │
+│       ▲                                   │        │
+│       │          ┌───────────────┐        │        │
+│       └──────────│  Translation  │◀───────┘        │
+│                  │ Codex→OpenAI  │  SSE stream     │
+│                  └───────────────┘                  │
+│                                                     │
+│  ┌──────────┐  ┌───────────────┐  ┌─────────────┐  │
+│  │   Auth   │  │  Fingerprint  │  │   Session   │  │
+│  │ OAuth/JWT│  │  Headers/UA   │  │   Manager   │  │
+│  └──────────┘  └───────────────┘  └─────────────┘  │
+│                                                     │
+└─────────────────────────────────────────────────────┘
+                         │
+                    curl subprocess
+                    (Chrome TLS)
+                         │
+                         ▼
+                    chatgpt.com
+              /backend-api/codex/responses
 ```
 
-## Configuration
-
-All configuration is in `config/default.yaml`:
-
-| Section | Key Settings |
-|---------|-------------|
-| `api` | `base_url`, `timeout_seconds` |
-| `client` | `originator`, `app_version`, `platform`, `arch` |
-| `model` | `default` model, `default_reasoning_effort` |
-| `auth` | `oauth_client_id`, `rotation_strategy`, `rate_limit_backoff_seconds` |
-| `server` | `host`, `port`, `proxy_api_key` |
+## 📦 可用模型 (Available Models)
 
-Environment variable overrides:
+| 模型 ID | 别名 | 说明 |
+|---------|------|------|
+| `gpt-5.2-codex` | `codex` | 最新 agentic 编程模型（默认） |
+| `gpt-5.1-codex-mini` | `codex-mini` | 轻量快��编程模型 |
 
-| Variable | Overrides |
-|----------|-----------|
-| `PORT` | `server.port` |
-| `CODEX_PLATFORM` | `client.platform` |
-| `CODEX_ARCH` | `client.arch` |
-| `CODEX_JWT_TOKEN` | `auth.jwt_token` |
+> 模型列表会随 Codex Desktop 版本更新自动同步。
 
-## Client Setup Examples
+## 🔗 客户端接入 (Client Setup)
 
 ### Cursor
 
-Settings > Models > OpenAI API Base:
+Settings → Models → OpenAI API Base:
 ```
 http://localhost:8080/v1
 ```
 
+API Key（从控制面板获取）:
+```
+codex-proxy-xxxxx
+```
+
 ### Continue (VS Code)
 
 `~/.continue/config.json`:
@@ -146,22 +151,115 @@ http://localhost:8080/v1
     "title": "Codex",
     "provider": "openai",
     "model": "codex",
-    "apiBase": "http://localhost:8080/v1"
+    "apiBase": "http://localhost:8080/v1",
+    "apiKey": "codex-proxy-xxxxx"
   }]
 }
 ```
 
-## Scripts
+### OpenAI Python SDK
+
+```python
+from openai import OpenAI
+
+client = OpenAI(
+    base_url="http://localhost:8080/v1",
+    api_key="codex-proxy-xxxxx"
+)
+
+response = client.chat.completions.create(
+    model="codex",
+    messages=[{"role": "user", "content": "Hello!"}],
+    stream=True
+)
+
+for chunk in response:
+    print(chunk.choices[0].delta.content or "", end="")
+```
+
+### OpenAI Node.js SDK
+
+```typescript
+import OpenAI from "openai";
+
+const client = new OpenAI({
+  baseURL: "http://localhost:8080/v1",
+  apiKey: "codex-proxy-xxxxx",
+});
+
+const stream = await client.chat.completions.create({
+  model: "codex",
+  messages: [{ role: "user", content: "Hello!" }],
+  stream: true,
+});
+
+for await (const chunk of stream) {
+  process.stdout.write(chunk.choices[0]?.delta?.content || "");
+}
+```
+
+## ⚙️ 配置说明 (Configuration)
+
+所有配置位于 `config/default.yaml`：
+
+| 分类 | 关键配置 | 说明 |
+|------|---------|------|
+| `server` | `host`, `port`, `proxy_api_key` | 服务监听地址与 API 密钥 |
+| `api` | `base_url`, `timeout_seconds` | 上游 API 地址与请求超时 |
+| `client_identity` | `app_version`, `build_number` | 模拟的 Codex Desktop 版本 |
+| `model` | `default`, `default_reasoning_effort` | 默认模型与推理强度 |
+| `auth` | `rotation_strategy`, `rate_limit_backoff_seconds` | 轮换策略与限流退避 |
+
+### 环境变量覆盖
+
+| 环境变量 | 覆盖配置 |
+|---------|---------|
+| `PORT` | `server.port` |
+| `CODEX_PLATFORM` | `client_identity.platform` |
+| `CODEX_ARCH` | `client_identity.arch` |
+
+## 📡 API 端点一览 (API Endpoints)
+
+| 端点 | 方法 | 说明 |
+|------|------|------|
+| `/v1/chat/completions` | POST | 聊天补全（核心端点） |
+| `/v1/models` | GET | 可用模型列表 |
+| `/health` | GET | 健康检查 |
+| `/auth/accounts` | GET | 账号列表与配额查询 |
+| `/auth/login` | GET | OAuth 登录入口 |
+| `/debug/fingerprint` | GET | 调试：查看当前伪装头信息 |
+
+## 🔧 命令 (Commands)
+
+| 命令 | 说明 |
+|------|------|
+| `npm run dev` | 开发模式启动（热重载） |
+| `npm run build` | 编译 TypeScript 到 `dist/` |
+| `npm start` | 运行编译后的生产版本 |
+
+## 📋 系统要求 (Requirements)
+
+- **Node.js** 18+
+- **curl** — 系统自带即可；安装 [curl-impersonate](https://github.com/lexiforest/curl-impersonate) 可获得完整 Chrome TLS 伪装
+- **ChatGPT 账号** — 普通账号即可
+
+## ⚠️ 注意事项 (Notes)
+
+- Codex API 为**流式输出专用**，设置 `stream: false` 时代理会内部流式收集后返回完整 JSON
+- 本项目依赖 Codex Desktop 的公开接口，上游版本更新可能导致接口变动
+- 建议在 **Linux / macOS** 上部署以获得完整 TLS 伪装能力（Windows 下 curl-impersonate 暂不可用）
+
+## 📄 许可协议 (License)
+
+本项目采用 **非商业许可 (Non-Commercial)**：
+
+- **允许**：个人学习、研究、自用部署
+- **禁止**：任何形式的商业用途，包括但不限于出售、转售、收费代理、商业产品集成
 
-| Command | Description |
-|---------|-------------|
-| `npm run dev` | Start dev server with hot reload |
-| `npm run build` | Compile TypeScript to `dist/` |
-| `npm start` | Run compiled server |
-| `npm run check-update` | Check for new Codex Desktop versions |
-| `npm run extract -- --path <asar>` | Extract fingerprint from Codex app |
-| `npm run apply-update` | Apply extracted fingerprint updates |
+本项目与 OpenAI 无关联。使用者需自行承担风险并遵守 OpenAI 的服务条款。
 
-## License
+---
 
-For personal use only. This project is not affiliated with OpenAI.
+<div align="center">
+  <sub>Built with Hono + TypeScript | Powered by Codex Desktop API</sub>
+</div>

README_EN.md

@@ -0,0 +1,265 @@
+<div align="center">
+
+  <h1>Codex Proxy</h1>
+  <h3>Your Local Codex Coding Assistant Gateway</h3>
+  <p>Expose Codex Desktop's capabilities as a standard OpenAI API, seamlessly connecting any AI client.</p>
+
+  <p>
+    <img src="https://img.shields.io/badge/Runtime-Node.js_18+-339933?style=flat-square&logo=nodedotjs&logoColor=white" alt="Node.js">
+    <img src="https://img.shields.io/badge/Language-TypeScript-3178C6?style=flat-square&logo=typescript&logoColor=white" alt="TypeScript">
+    <img src="https://img.shields.io/badge/Framework-Hono-E36002?style=flat-square" alt="Hono">
+    <img src="https://img.shields.io/badge/License-Non--Commercial-red?style=flat-square" alt="License">
+  </p>
+
+  <p>
+    <a href="#-quick-start">Quick Start</a> •
+    <a href="#-features">Features</a> •
+    <a href="#-architecture">Architecture</a> •
+    <a href="#-client-setup">Client Setup</a> •
+    <a href="#-configuration">Configuration</a>
+  </p>
+
+  <p>
+    <a href="./README.md">简体中文</a> |
+    <strong>English</strong>
+  </p>
+
+</div>
+
+---
+
+**Codex Proxy** is a lightweight local gateway that translates the [Codex Desktop](https://openai.com/codex) Responses API into a standard OpenAI-compatible `/v1/chat/completions` endpoint. Use Codex coding models directly in Cursor, Continue, VS Code, or any OpenAI-compatible client.
+
+Just a ChatGPT account and this proxy — your own personal AI coding assistant gateway, running locally.
+
+## 🚀 Quick Start
+
+```bash
+# 1. Clone the repo
+git clone https://github.com/icebear0828/codex-proxy.git
+cd codex-proxy
+
+# 2. Install dependencies
+npm install
+
+# 3. Start the proxy (dev mode with hot reload)
+npm run dev
+
+# 4. Open the dashboard and log in with your ChatGPT account
+#    http://localhost:8080
+
+# 5. Test a request
+curl http://localhost:8080/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "codex",
+    "messages": [{"role": "user", "content": "Hello!"}],
+    "stream": true
+  }'
+```
+
+## 🌟 Features
+
+### 1. 🔌 Full Protocol Compatibility
+- Complete `/v1/chat/completions` and `/v1/models` endpoint support
+- SSE streaming output, works with all OpenAI SDKs and clients
+- Automatic bidirectional translation between Chat Completions and Codex Responses API
+
+### 2. 🔐 Account Management & Smart Rotation
+- **OAuth PKCE login** — one-click browser auth, no manual token copying
+- **Multi-account rotation** — `least_used` and `round_robin` scheduling strategies
+- **Auto token refresh** — JWT renewed automatically before expiry
+- **Real-time quota monitoring** — dashboard shows remaining usage per account
+
+### 3. 🛡️ Anti-Detection & Protocol Impersonation
+- **Chrome TLS fingerprint** — curl-impersonate replicates the full Chrome 136 TLS handshake
+- **Desktop header replication** — `originator`, `User-Agent`, `sec-ch-*` headers in exact Codex Desktop order
+- **Desktop context injection** — every request includes the Codex Desktop system prompt for full feature parity
+- **Cookie persistence** — automatic Cloudflare cookie capture and replay
+- **Timing jitter** — randomized delays on scheduled operations to eliminate mechanical patterns
+
+### 4. 🔄 Session & Version Management
+- **Multi-turn conversations** — automatic `previous_response_id` for context continuity
+- **Appcast version tracking** — polls Codex Desktop update feed, auto-syncs `app_version` and `build_number`
+- **Web dashboard** — account management, usage monitoring, and status overview in one place
+
+## 🏗️ Architecture
+
+```
+                            Codex Proxy
+┌─────────────────────────────────────────────────────┐
+│                                                     │
+│  Client (Cursor / Continue / SDK)                   │
+│       │                                             │
+│  POST /v1/chat/completions                          │
+│       │                                             │
+│       ▼                                             │
+│  ┌──────────┐    ┌───────────────┐    ┌──────────┐  │
+│  │  Routes   │──▶│  Translation  │──▶│  Proxy   │  │
+│  │  (Hono)  │   │ OpenAI→Codex  │   │ curl TLS │  │
+│  └──────────┘   └───────────────┘   └────┬─────┘  │
+│       ▲                                   │        │
+│       │          ┌───────────────┐        │        │
+│       └──────────│  Translation  │◀───────┘        │
+│                  │ Codex→OpenAI  │  SSE stream     │
+│                  └───────────────┘                  │
+│                                                     │
+│  ┌──────────┐  ┌───────────────┐  ┌─────────────┐  │
+│  │   Auth   │  │  Fingerprint  │  │   Session   │  │
+│  │ OAuth/JWT│  │  Headers/UA   │  │   Manager   │  │
+│  └──────────┘  └───────────────┘  └─────────────┘  │
+│                                                     │
+└─────────────────────────────────────────────────────┘
+                         │
+                    curl subprocess
+                    (Chrome TLS)
+                         │
+                         ▼
+                    chatgpt.com
+              /backend-api/codex/responses
+```
+
+## 📦 Available Models
+
+| Model ID | Alias | Description |
+|----------|-------|-------------|
+| `gpt-5.2-codex` | `codex` | Latest agentic coding model (default) |
+| `gpt-5.1-codex-mini` | `codex-mini` | Lightweight, fast coding model |
+
+> Models are automatically synced when new Codex Desktop versions are released.
+
+## 🔗 Client Setup
+
+### Cursor
+
+Settings → Models → OpenAI API Base:
+```
+http://localhost:8080/v1
+```
+
+API Key (from the dashboard):
+```
+codex-proxy-xxxxx
+```
+
+### Continue (VS Code)
+
+`~/.continue/config.json`:
+```json
+{
+  "models": [{
+    "title": "Codex",
+    "provider": "openai",
+    "model": "codex",
+    "apiBase": "http://localhost:8080/v1",
+    "apiKey": "codex-proxy-xxxxx"
+  }]
+}
+```
+
+### OpenAI Python SDK
+
+```python
+from openai import OpenAI
+
+client = OpenAI(
+    base_url="http://localhost:8080/v1",
+    api_key="codex-proxy-xxxxx"
+)
+
+response = client.chat.completions.create(
+    model="codex",
+    messages=[{"role": "user", "content": "Hello!"}],
+    stream=True
+)
+
+for chunk in response:
+    print(chunk.choices[0].delta.content or "", end="")
+```
+
+### OpenAI Node.js SDK
+
+```typescript
+import OpenAI from "openai";
+
+const client = new OpenAI({
+  baseURL: "http://localhost:8080/v1",
+  apiKey: "codex-proxy-xxxxx",
+});
+
+const stream = await client.chat.completions.create({
+  model: "codex",
+  messages: [{ role: "user", content: "Hello!" }],
+  stream: true,
+});
+
+for await (const chunk of stream) {
+  process.stdout.write(chunk.choices[0]?.delta?.content || "");
+}
+```
+
+## ⚙️ Configuration
+
+All configuration is in `config/default.yaml`:
+
+| Section | Key Settings | Description |
+|---------|-------------|-------------|
+| `server` | `host`, `port`, `proxy_api_key` | Listen address and API key |
+| `api` | `base_url`, `timeout_seconds` | Upstream API URL and timeout |
+| `client_identity` | `app_version`, `build_number` | Codex Desktop version to impersonate |
+| `model` | `default`, `default_reasoning_effort` | Default model and reasoning effort |
+| `auth` | `rotation_strategy`, `rate_limit_backoff_seconds` | Rotation strategy and rate limit backoff |
+
+### Environment Variable Overrides
+
+| Variable | Overrides |
+|----------|-----------|
+| `PORT` | `server.port` |
+| `CODEX_PLATFORM` | `client_identity.platform` |
+| `CODEX_ARCH` | `client_identity.arch` |
+
+## 📡 API Endpoints
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `/v1/chat/completions` | POST | Chat completions (main endpoint) |
+| `/v1/models` | GET | List available models |
+| `/health` | GET | Health check |
+| `/auth/accounts` | GET | Account list and quota |
+| `/auth/login` | GET | OAuth login entry |
+| `/debug/fingerprint` | GET | Debug: view current impersonation headers |
+
+## 🔧 Commands
+
+| Command | Description |
+|---------|-------------|
+| `npm run dev` | Start dev server with hot reload |
+| `npm run build` | Compile TypeScript to `dist/` |
+| `npm start` | Run compiled production server |
+
+## 📋 Requirements
+
+- **Node.js** 18+
+- **curl** — system curl works out of the box; install [curl-impersonate](https://github.com/lexiforest/curl-impersonate) for full Chrome TLS fingerprinting
+- **ChatGPT account** — standard account is sufficient
+
+## ⚠️ Notes
+
+- The Codex API is **stream-only**. When `stream: false` is set, the proxy streams internally and returns the assembled response as a single JSON object.
+- This project relies on Codex Desktop's public API. Upstream version updates may cause breaking changes.
+- Deploy on **Linux / macOS** for full TLS impersonation. On Windows, curl-impersonate is not available and the proxy falls back to system curl.
+
+## 📄 License
+
+This project is licensed under **Non-Commercial** terms:
+
+- **Allowed**: Personal learning, research, self-hosted deployment
+- **Prohibited**: Any commercial use, including but not limited to selling, reselling, paid proxy services, or integration into commercial products
+
+This project is not affiliated with OpenAI. Users assume all risks and must comply with OpenAI's Terms of Service.
+
+---
+
+<div align="center">
+  <sub>Built with Hono + TypeScript | Powered by Codex Desktop API</sub>
+</div>