Codex with ChatGPT v0.1.2 发布：增强 MCP 结构化输出与 Windows 稳定性

Official release notes (https://github.com/XiaoDuoYa/codex-with-chatgpt/releases/tag/v0.1.2) Codex with ChatGPT v0.1.2 ChatGPT thinks. Codex works. v0.1.2 is a reliability and interoperability update focused on making the bridge safer to start, quieter on Windows, and easier for MCP clients to consume correctly. The core model remains unchanged: ChatGPT plans and reviews; Codex owns execution. No API key. No reverse proxy. No cookies or session extraction. ✨ Highlights - Structured MCP outputs - Read-only MCP tools now advertise structured output schemas - Successful calls return matching "structuredContent" while keeping the existing text JSON content for compatibility - Workspace, file, search, Git, test status, execution summary, and execution output results are covered - This makes C2C easier and safer for MCP clients to consume without guessing response shapes - Structured-output hardening - "c2c record" now validates numeric arguments before saving them - Execution records are validated on write - Malformed historical records are skipped safely on read instead of breaking otherwise read-only MCP calls - "test_status" and "execution_summary" now stay aligned with their advertised schemas - Workspace and package metadata returned by "workspace_info" are sanitized before entering structured output - "git_diff" documentation now uses the real pagination fields: "hasMore" and "nextOffset" - Safer Windows Quick Tunnel startup - Quick Tunnel startup now fails closed instead of reporting success too early - Startup completes only after the public "/health" endpoint identifies a healthy C2C Bridge - Invalid or reserved "api.trycloudflare.com" URLs are rejected - Early "cloudflared" exits, invalid health responses, timeouts, and stopped startup are handled as failures - "C2C_CLOUDFLARED_PATH" can explicitly select a custom "cloudflared" executable - A quieter Windows experience - Bridge daemon processes no longer open unnecessary console windows - Named Tunnel processes no longer flash console windows - Quick Tunnel processes are also hidden - C2C can now stay in the background without repeatedly popping terminal windows during normal Windows use 🛠 Reliability & UX - "doctor" no longer assumes an inconclusive local bridge probe means the bridge process is dead - If the process still exists, C2C will not start a second bridge or unnecessarily ask ChatGPT to delete the connector - Quick Tunnel URL discovery and public-health verification are stricter - Windows tunnel process handling is more predictable - Several setup and troubleshooting wording issues were cleaned up Idea for the bridge-probe reliability improvement from @AtlaxTech (#43). Windows background-process improvements include contributions from @moonjun (#164). 🔐 Safety & correctness - Structured MCP results now have explicit schemas instead of relying only on text conventions - Malformed persisted execution data cannot silently violate the MCP output contract - Invalid numeric execution metadata is rejected before persistence - Unexpected values in ".c2c.json" and "package.json" are filtered before being exposed through "workspace_info" - Quick Tunnel startup requires confirmation from the actual public C2C "/health" endpoint - No new MCP permissions were introduced by these changes - No reverse proxy, cookie, session-extraction, or credential-sharing behavior was added 🧪 Tests Regression coverage was expanded for: - structured MCP output schemas - malformed persisted execution records - invalid "c2c record" CLI arguments - workspace metadata sanitization - bridge runtime detection - Windows Quick Tunnel startup and failure handling The structured-output hardening follow-up reports 164 tests passed, along with successful typecheck, build, and "git diff --check". 📚 Documentation - Fixed "login" / "log in" wording in the Skill and README - Updated Quick Tunnel troubleshooting guidance - Corrected "git_diff" pagination documentation to use "hasMore" / "nextOffset" - Documentation now better reflects the stricter bridge and tunnel behavior 🙌 Contributors Thanks to everyone who reported issues, tested C2C across different environments, and contributed fixes or ideas. Special thanks to: - @AtlaxTech — bridge-process diagnosis / reliability idea - @moonjun — Windows background-process improvements - @peak-xiong — structured MCP output schemas and hardening - Contributors behind the documentation fixes and Windows tunnel testing --- 中文 这是 v0.1.2，重点是把 C2C 的稳定性、Windows 使用体验和 MCP 输出规范继续打磨扎实。 核心理念仍然没变： «ChatGPT 负责思考，Codex 负责干活。» 这版重点 - MCP 正式加入 structured output schema - 只读 MCP 工具现在明确告诉客户端自己会返回什么格式 - 同时保留原来的 JSON 文本输出，兼容已有使用方式 - workspace、文件、搜索、Git、测试状态、execution summary / output 等结果都覆盖到了 - structured output 二次加固 - "c2c record" 的数字输入会先检查再保存 - execution record 写入前验证 - 历史脏数据不会再把只读 MCP 工具直接搞崩，而是安全跳过 - "test_status"、"execution_summary" 与 schema 保持一致 - "workspace_info" 会过滤异常的项目配置和 package metadata - "git_diff" 分页说明修正为实际使用的 "hasMore" / "nextOffset" - Windows Quick Tunnel 更可靠 - 不再 tunnel 一启动就默认“成功” - 必须真正从公网 "/health" 确认这是正常运行的 C2C Bridge 才算启动完成 - 拒绝错误的 "api.trycloudflare.com" 地址 - cloudflared 提前退出、health 异常、启动超时等都会正确失败 - 支持用 "C2C_CLOUDFLARED_PATH" 指定 cloudflared - Windows 后台终于更安静 - Bridge daemon 不再弹黑色控制台窗口 - Named Tunnel 不再弹 - Quick Tunnel 也不再弹 - 正常使用 C2C 时后台进程不会疯狂闪终端窗口 - doctor 更聪明 - 本地 probe 一时没有得到明确结果，不再直接认定 bridge 死了 - 只要原进程还存在，就不会再启动第二个 bridge - 也不会因为一次模糊检测就让 ChatGPT 去删 connector 安全与正确性 这次没有扩大 MCP 权限，也没有加入 reverse proxy、cookie、session 提取之类的新行为。 主要是在原来的架构上继续收紧： 输入更严格、输出更确定、脏数据更难把系统搞崩、Tunnel 启动成功的判断更可信。 另外补了更多 regression tests；最新 structured-output 加固相关测试达到 164 tests passed。 感谢 @AtlaxTech、@moonjun、@peak-xiong 以及所有提交 Issue、测试 Windows / Tunnel 场景、修正文档和贡献代码的人。❤️ 最重要的一点仍然是： 能让 Agent 自己完成的事情，就不要让用户完成。 Repository README (current repository documentation, not release changes) # Codex with ChatGPT > ChatGPT thinks. Codex works. > ChatGPT 负责思考，Codex 负责干活。 > [!IMPORTANT] > **遇到问题？** 请先向 Codex 发送 **「更新 Codex with ChatGPT」** 并重试。更新到最新版本可以解决大多数已知问题。 > **Having trouble?** First ask Codex to **“Update Codex with ChatGPT”** and try again. Updating to the latest version resolves most known issues. ## The problem · 解决什么问题 **中文** — ChatGPT 付费订阅的网页版额度大量闲置，Codex 却在消耗紧张的 API 额度做规划和 Review。本项目把"思考"交给你已付费的网页版 ChatGPT， Codex 只负责执行。不用 API Key、不搞逆向代理——官方网页 + 只读 MCP 桥接。 **EN** — ChatGPT Plus/Pro web quota sits idle while your coding agent burns scarce API/Codex tokens on planning and review. This project moves the thinking to the subscription you already pay for; Codex only executes. No API keys, no reverse proxy — official web UI plus a read-only MCP bridge. ## What it is · 这是什么 **中文** — 把 ChatGPT 网页版变成 Codex 编码会话的"规划与审查大脑"，执行权 完全保留在 Codex 手里。你的仓库永远不会被上传：ChatGPT 通过一条安全的、 OAuth 保护的**只读** MCP 连接，按需读取当前工作区里它真正需要的那几行代码。 **EN** — Use the ChatGPT web app as the planning and review brain for your Codex coding sessions, while Codex keeps full ownership of execution. Your repository is never uploaded: ChatGPT reads exactly the lines it needs through a secure, OAuth-protected, **read-only** MCP connection to your current workspace. Detailed docs below are in English · 详细中文文档见 **[README.zh-CN.md](README.zh-CN.md)** ## One-paste install · 一段话安装 **中文** — 不懂 git、Node、终端？完全不需要懂。把下面这段话原样复制给你的 编码 Agent（Codex），然后去倒杯咖啡： ```text 请帮我完整安装并配置 Codex with ChatGPT，全程自动，我是不懂技术的小白， 所有事情你自己做： 1. 环境自检：需要 git 和 Node.js ≥ 20，缺什么就自动安装 （macOS 用 Homebrew，Windows 用 winget），同时安装 cloudflared。 2. 下载：把 https://github.com/XiaoDuoYa/codex-with-chatgpt 克隆到 ~/codex-with-chatgpt（已存在就 git pull 更新）。 3. 构建：在该目录里执行 corepack pnpm install 和 corepack pnpm build。 4. 安装 Skill：把仓库里的 skill/SKILL.md 复制到 ~/.codex/skills/codex-with-chatgpt/SKILL.md，并把文件中 "The codex-with-chatgpt checkout lives at:" 那一行的路径改成实际克隆路径。 5. 首次配置：按 SKILL.md 里的 first-time setup 流程执行 （运行 c2c setup，用内置浏览器打开 ChatGPT 配置连接器并输入配对码）。 全程只用内置浏览器，禁止打开任何第三方浏览器。 6. 只有遇到需要我登录（ChatGPT / Cloudflare）、验证码或两步验证时才叫我， 而且一次只告诉我一个动作。 7. 完成后给我看 ✓ 清单，并确认文件读取测试通过。我不懂 MCP、OAuth、 Tunnel、端口这些词，不要向我解释；出了问题先自己修。 ``` **EN** — Don't know git, Node, or terminals? You don't need to. Copy the paragraph below, paste it to your coding agent (Codex), and go grab a coffee: ```text Please install and configure "Codex with ChatGPT" for me, fully automatically. I am a non-technical user — do everything yourself: 1. Check the environment: git and Node.js >= 20 must be available. Install anything missing yourself (macOS: Homebrew, Windows: winget). Also install cloudflared. 2. Download: clone https://github.com/XiaoDuoYa/codex-with-chatgpt into ~/codex-with-chatgpt (if it already exists, git pull to update). 3. Build: inside that folder run `corepack pnpm install` then `corepack pnpm build`. 4. Install the Skill: copy skill/SKILL.md to ~/.codex/skills/codex-with-chatgpt/SKILL.md, and update the line "The codex-with-chatgpt checkout lives at:" to the actual clone path. 5. First-time setup: follow the SKILL.md "first-time setup" workflow (run c2c setup, configure the ChatGPT connector in the BUILT-IN browser, enter the pairing code). Never open a third-party browser. 6. Only interrupt me for logins (ChatGPT / Cloudflare), CAPTCHAs or 2FA — and give me exactly ONE action at a time. 7. When done, show me the ✓ checklist and confirm the file-read test passed. I don't know what MCP, OAuth, tunnels or ports are. Don't explain them. If anything breaks, fix it yourself first. ``` **Updates · 更新** — The Skill checks GitHub once a day and updates itself when a new version is released; no action needed. You can also say "更新 Codex with ChatGPT" anytime. / Skill 每天自动检查一次 GitHub，有新版本会自动更新，无需任何操作； 也可以随时对 Codex 说"更新 Codex with ChatGPT"。 --- *The sections below are in English. 以下详细内容为英文，中文完整版见 [README.zh-CN.md](README.zh-CN.md)。* ## Install → Setup → Use (manual) 1. Install the Codex Skill: copy `skill/` to `~/.codex/skills/codex-with-chatgpt/`. 2. Tell Codex: **"Set up Codex with ChatGPT."** (中文: "使用 Codex with ChatGPT 完成首次配置。") 3. Use Codex normally: **"Use Codex with ChatGPT to implement XXX."** That's the whole manual. You don't need to know what MCP, OAuth, tunnels, ports or localhost are — Codex configures everything automatically and you just see: ``` Codex with ChatGPT ✓ Project detected ✓ Workspace Bridge started ✓ Secure connection established ✓ ChatGPT connected ✓ File read test passed Ready. ``` The only steps that may need you: logging into ChatGPT (and, if you want a stable hostname, logging into Cloudflare once). A **new** workspace also asks you to create a ChatGPT Project (collection) once — pick **project-only memory**, name it after the workspace. If the sidebar has no Projects row, hover **Chats**, open the … menu, and choose **Organize by project**. Codex then saves that collection link and starts chats from that page. Existing workspaces that already have a C2C chat stay on the old one-conversation style until you ask to switch. ### Optional stable hostname The default public address is a temporary Cloudflare URL. It changes when the bridge restarts, and Codex repairs ChatGPT by deleting that workspace's connector and adding it again. If you have a Cloudflare account and a domain already on Cloudflare, first-time setup (and the next coding session, once) will ask whether you want a stable hostname such as `c2c-.your-domain.com`. That path opens a browser so you can authorize Cloudflare. After that, the ChatGPT connector keeps working across restarts. If you skip it, or the login fails, Codex stays on the temporary address — same features, just a slower repair. Credentials stay in the OS app state directory, not in the project. ## How it works ``` ┌───────────────────────────┐ │ ChatGPT Web │ │ Reason / Plan / Review │ └──────────┬──────────▲─────┘ │ │ MCP │ │ Computer Use Data Plane │ │ Control Plane ( dist/, exposes the `c2c` bin pnpm test # vitest: 146 tests (path security, OAuth, pairing, MCP e2e) c2c setup # bridge + tunnel + pairing code, all in one c2c sandbox-allow # whitelist the settings dir in Codex (macOS + Windows) c2c status / doctor / pair / unpair / logs / stop ``` Requirements: Node.js >= 20, git. `cloudflared` for the public connection (auto-detected; the Skill installs it for you). Docs: [architecture](docs/architecture.md) · [protocol](docs/protocol.md) · [security](docs/security.md) · [troubleshooting](docs/troubleshooting.md) ## Project layout ``` src/ bridge/ loopback HTTP server, port recovery, admin API mcp/ 9 read-only tools, stateless Streamable HTTP auth/ OAuth 2.1 (PKCE, DCR, refresh rotation, revocation) pairing/ one-time pairing codes (CSPRNG, TTL, rate limits) workspace/ path containment, sensitive-file policy, search, git tunnel/ TunnelProvider abstraction + Cloudflare Quick/Named Tunnel execution/ execution records for the review loop process/ daemon lifecycle cli/ the c2c CLI skill/ the Codex Skill (the real UX layer) tests/ unit + integration tests docs/ architecture / protocol / security / troubleshooting ``` ## Status & disclaimer V1. Verified end-to-end: bridge, OAuth + pairing, public tunnel, ChatGPT connector setup, zero-touch first-run experience. **Unofficial community project. Not affiliated with or endorsed by OpenAI.** ## License [MIT](LICENSE)