2836048681 GitHub avatar

cc-switch-codexcont

2836048681

A personal enhanced build of CC Switch v3.17.0 integrating CodexCont for automatic Codex reasoning continuation and Grok Build for independent Grok provider management and routing.

Stars

29

7-day growth

No data

Forks

4

Open issues

0

License

MIT

Last updated

2026-07-28

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It embeds two previously separate features—Codex reasoning continuation and Grok Build configuration—directly into CC Switch's UI and routing pipeline, eliminating the need for extra forwarding services.

Who it is for

  • Existing CC Switch users who want Codex auto-continuation
  • Grok/xAI users needing provider switching and routing within CC Switch
  • Windows users seeking an all-in-one proxy solution for Codex and Grok
  • Developers familiar with CC Switch who want persistent UI-enabled toggles

Use cases

  • Enable Codex reasoning auto-continuation via a UI toggle without editing config files
  • Manage multiple Grok providers (xAI, compatible) with independent routing and failover
  • Route both /v1/responses and /v1/chat/completions requests through the same CC Switch instance
  • Persist CodexCont settings in the database and override them with environment variables

Strengths

  • Retains all upstream CC Switch v3.17.0 features (snapshot, provider routing, failover, subscriptions) intact
  • Adds a UI switch for CodexCont that is persisted via database schema v14 upgrade
  • Codex continuation reuses CC Switch's RequestForwarder with retry/failover, not bypassing routing or usage tracking
  • Grok Build is a fully separate top-level application with its own providers, profiles, and backup mechanism

Considerations

  • Only Windows builds are provided (Setup, MSI, Portable); no macOS or Linux support mentioned
  • This is a personal fork, not the official CC Switch repository, and may not track upstream updates
  • Users must still configure their own Codex/Grok providers upstream; the build does not include any default API keys

README quick start

CC Switch CodexCont + Grok Build

基于 CC Switch v3.17.0 的 Codex reasoning 自动续写与 Grok Build 增强版

下载 Release · 上游 CC Switch · CodexCont · Grok Build Switch

这个仓库是什么

这是一个基于 CC Switch v3.17.0 发布提交 3d176b9 的个人增强构建。

本构建保留 CC Switch v3.17.0 原有的项目快照、供应商管理、代理路由、故障转移、官方 ChatGPT 订阅代理、原生 Anthropic Messages,以及 Codex / Claude / Gemini / OpenCode / OpenClaw / Hermes 配置管理能力,并额外加入两个功能:

  • CodexCont 风格的 Codex reasoning 自动续写
  • 独立的 Grok Build 供应商、配置管理和代理路由

核心链路保持为:

Codex -> CC Switch 本地代理 -> 多上游路由 / 故障转移 -> 模型服务
Grok  -> CC Switch Grok 路由 -> xAI / Grok 兼容上游

Codex 续写逻辑和 Grok 路由都集成在 CC Switch 内部,不需要额外运行独立转发服务。

相对上游 CC Switch 的新增改动

1. 新增 CodexCont UI 开关

在设置页加入独立开关:

设置 -> 代理 -> Codex reasoning 自动续写(CodexCont)

可以在 UI 中控制是否启用 Codex reasoning 自动续写,不需要手动修改配置文件。

2. 支持持久化 CodexCont 配置

新增数据库配置项:

codex_continue_config

配置由 CC Switch 自身保存和读取,重启后继续生效。

3. /v1/responses 自动续写处理

当 Codex 请求经过 CC Switch 的 /v1/responses 时,后端会读取 CodexCont 配置,并在满足条件时自动执行 reasoning continuation。

续写请求仍然使用 CC Switch 原有的 provider 转发和 retry/failover 流程:

RequestForwarder::forward_with_retry(... providers)

因此不会绕开 CC Switch 的路由、故障转移和用量记录。

4. 保留 CC Switch 路由和故障转移体验

本构建没有把 Codex provider 固定到单独服务,也不要求把 Codex 指向额外端口。推荐配置仍然是 CC Switch 本地代理,例如:

base_url = "http://127.0.0.1:15721/v1"

5. native Responses provider 安全门控

只有候选 providers 都使用原生 Responses 链路、不需要 Responses -> Chat Completions 转换时,才启用 Codex reasoning continuation。

这样可以避免把 reasoning.encrypted_content 注入不兼容的 Chat provider。

6. CodexCont 环境变量覆盖

除 UI 配置外,还支持环境变量覆盖运行参数:

CCSWITCH_CODEX_CONTINUE
CCSWITCH_CODEX_CONTINUE_MAX
CCSWITCH_CODEX_CONTINUE_STEP
CCSWITCH_CODEX_CONTINUE_MARKER

7. 独立 Grok Build 应用

Grok 作为独立顶层应用

Description

CC Switch 3.16.5 with CodexCont integration and Windows release assets

Related repositories

Similar projects matched by category, topics, and programming language.

l0ng-ai
Featured
l0ng-ai GitHub avatar

tty7

tty7 is a fast, GPU-accelerated terminal workbench with persistent sessions, built-in input tools, SSH support, and coding agent awareness.

Developer ToolsCLI & Terminal
359
steelbrain
Featured
steelbrain GitHub avatar

reims-vgpu

reims-vgpu is an experimental virtual GPU for macOS guests that uses QEMU to decode the guest's GPU command stream and execute it through Metal or Vulkan, leveraging the built-in AppleParavirtGPU driver without requiring custom kexts.

Desktop Apps
143
m-novotny
Featured
m-novotny GitHub avatar

memguard-rs

A Rust library that provides secure memory handling primitives including zeroization on drop, memory locking, constant-time comparison, and compile-time guarded regions, with zero dependencies and no_std support.

Embedded & IoTSecurity
131