dsh-codex-subscription 2.2.10 → 2.3.0

package/README.en.md

@@ -16,12 +16,21 @@ No OpenAI API key or Codex CLI. Models, search, quota, and image generation stay
 [![MIT](https://img.shields.io/badge/license-MIT-111111.svg)](LICENSE)
 [![Star](https://img.shields.io/github/stars/WSL043/dsh-codex-subscription?style=flat&logo=github&label=Star)](https://github.com/WSL043/dsh-codex-subscription/stargazers)
 
-[Features](#feature-overview) · [Install](#install) · [Daily use](#inside-the-plugin) · [Examples](#from-sketch-to-image) · [User guide](docs/GUIDE.en.md) · [Update and uninstall](#update-and-uninstall)
+[What's new](#whats-new-in-230) · [Features](#feature-overview) · [Install](#install) · [Daily use](#inside-the-plugin) · [Examples](#from-sketch-to-image) · [User guide](docs/GUIDE.en.md) · [Update and uninstall](#update-and-uninstall)
 
 </div>
 
 <p align="center"><img src="docs/assets/product-account-demo-en.png" width="1100" alt="Codex subscription account and quota interface with model choices, quota windows and alerts; demonstration data"></p>
 
+## What's new in 2.3.0
+
+- **The official desktop app works.** The official DeepSeek Harness desktop app (0.2.0-rc.2) passed a hands-on acceptance run: plugin install, sign-in, model list, hiding models, a chat with a tool call, and image generation. See [Install](#install).
+- **Models come only from your account.** The model list is read entirely from the account catalog; before sign-in or when the read fails no model is shown, and there is no built-in fallback. Under Models & runtime you can show or hide each model in the composer.
+- **Settings reorganised.** Models and search are separate cards, switches use DSH's own control, and dead buttons no longer appear while signed out.
+- **Ultrafast speed tier (Beta).** The new tier from OpenAI's developer event. It appears in the model menu only when your account catalog lists it for that model (today Pro 500 with GPT-6 Astra); other accounts never see it and it is never silently swapped for another tier. It is Beta because it has not been run on a Pro 500 account yet.
+- **Shorter support diagnostics.** "Copy report" now copies the conclusions, recent failures and per-feature status (about 1 KB); the full report can still be copied separately from the details.
+- **Subtasks and stability.** Codex subtasks now really take effect on 0.2.0 and no longer expose the apps connected to your account; image generation being rejected by the host on the desktop app is fixed (2.2.11).
+
 ## Feature overview
 
 Subscription access, model controls, quota management, and image creation in one DSH workflow. Each capability is listed below; model access and quota depend on your account.
@@ -34,7 +43,7 @@ Subscription access, model controls, quota management, and image creation in one
 | **Model catalog sync** | Read available models and reasoning levels from your account, with manual refresh; models marked for retirement show their retirement date next to the model name, without automatically switching your selection | New models appear when your account offers them; the account catalog is the only source, so signed out or when it cannot be read no models are shown and no bundled list stands in |
 | **Model list** | Show or hide individual models in the composer picker under Models & runtime | Display only; selected models and requests still work |
 | **Reasoning levels** | Choose supported reasoning effort in the composer | Levels vary by model |
-| **Fast mode** | Switch supported models between Standard and Fast, with a lightning indicator | Standard by default; Fast increases usage |
+| **Speed tiers** | Switch supported models between Standard and Fast; Ultrafast (Beta) too when the account catalog offers it, with a lightning indicator | Standard by default; Fast and Ultrafast increase usage, and Ultrafast is Pro 500 only today |
 | **Response verbosity** | Request short, medium, or detailed responses | Follows the model default unless changed |
 | **Input image detail** | Choose Low, High, Original, or Auto detail for images sent to the chat model | Affects input tokens; Auto by default |
 | **Subscription web search** | Use Codex search or choose DSH search | Auto, DSH, and Codex routing options |
@@ -81,7 +90,7 @@ Subscription access, model controls, quota management, and image creation in one
 | **Subtask component management** | Install, disable, or uninstall the official Codex subtask component from settings | Managed by supported DSH hosts; no extra login |
 | **Cloud context compaction** | Continue long conversations using Codex compaction while retaining DSH history and native compaction | Beta / experimental; off by default, enabled requests use SSE |
 | **WebSocket transport** | Experimentally reuse connections and context transfers, with SSE fallback on connection failure | Beta / experimental; SSE by default, no speed guarantee |
-| **Support diagnostics** | Inspect prerequisites, recent operation results and host registrations without credentials or conversation text | Settings → Maintenance; unverified does not mean broken |
+| **Support diagnostics** | Inspect prerequisites and recent operation results, and copy a short report (about 1 KB; the full one can be copied separately) without credentials or conversation text | Settings → Maintenance; unverified does not mean broken |
 | **Forecast cache cleanup** | Inspect and clear local quota forecast history | Keeps sign-in, drafts, originals, and shared dependencies |
 
 [Usage details, limitations, and optional components →](docs/GUIDE.en.md)
@@ -101,16 +110,16 @@ This plugin supports the latest DeepSeek Harness release recorded in its package
 ### Install from the Plugins page (recommended)
 
 1. Open **Plugins → Add plugin** in DSH.
-2. Paste this package name into the **Package name or address** field:
+2. Paste this package name, with its version, into the **Package name or address** field:
 
    ```text
-   dsh-codex-subscription
+   dsh-codex-subscription@2.3.0
    ```
 
 3. Click **Install** and wait for completion. Follow the page instructions; save your work before restarting if requested.
 4. Open **Settings → Codex**, sign in to ChatGPT, then select a Codex model in your conversation.
 
-The package name installs the latest stable release. For a specific version, enter `dsh-codex-subscription@version`; find version numbers on the [releases page](https://github.com/WSL043/dsh-codex-subscription/releases).
+Why the version is included: DSH's plugin page only installs versions that are at least 24 hours old, so the bare package name can install an older release during the first day after a new one, and that older release may not support your DSH. With the version you get exactly this release; a day after release the bare `dsh-codex-subscription` also resolves to the latest stable release. Other version numbers are on the [releases page](https://github.com/WSL043/dsh-codex-subscription/releases).
 
 <details>
 <summary>Terminal installation (with an existing dsh command)</summary>
@@ -139,7 +148,7 @@ dsh --profile headless "Reply with only the word: ok"
 
 ### Everyday controls, right in the composer
 
-Choose a model, adjust reasoning, enable Fast mode, and check remaining quota without opening settings for every request.
+Choose a model, adjust reasoning, switch speed tiers, and check remaining quota without opening settings for every request.
 
 <p align="center"><img src="docs/assets/composer-quota-en.png" width="800" alt="DSH composer with GPT-6-Luna, reasoning controls and remaining quota"></p>
 
@@ -273,9 +282,9 @@ dsh plugin --profile web remove dsh-codex-subscription
 ## Troubleshooting
 
 <details>
-<summary>DSH 0.2.0-rc.2 reports plugin 2.2.4 as incompatible?</summary>
+<summary>DSH 0.2.0-rc.2 reports the plugin as incompatible?</summary>
 
-DSH 0.2.0-rc.2 did not exist when 2.2.4 and earlier were published, and a published package's compatibility declaration cannot be changed. Update to 2.2.5 or later from the **Plugins** page, or run `dsh plugin --profile web update dsh-codex-subscription`, then restart DSH.
+DSH 0.2.0-rc.2 did not exist when any release before 2.2.10 was published, and a published package's compatibility declaration cannot be changed. Install 2.2.10 or later (a bare package name can resolve to an older release during the first 24 hours, so enter `dsh-codex-subscription@2.2.10`) from the **Plugins** page, or run `dsh plugin --profile web update dsh-codex-subscription`, then restart DSH.
 
 </details>
 

package/README.md

@@ -16,12 +16,21 @@
 [![MIT](https://img.shields.io/badge/license-MIT-111111.svg)](LICENSE)
 [![Star](https://img.shields.io/github/stars/WSL043/dsh-codex-subscription?style=flat&logo=github&label=Star)](https://github.com/WSL043/dsh-codex-subscription/stargazers)
 
-[功能](#功能一览) · [安装](#安装) · [日常使用](#界面一览) · [作品案例](#从草图到作品) · [使用指南](docs/GUIDE.zh-CN.md) · [更新与卸载](#更新与卸载)
+[更新亮点](#230-更新亮点) · [功能](#功能一览) · [安装](#安装) · [日常使用](#界面一览) · [作品案例](#从草图到作品) · [使用指南](docs/GUIDE.zh-CN.md) · [更新与卸载](#更新与卸载)
 
 </div>
 
 <p align="center"><img src="docs/assets/product-account-demo.png" width="1100" alt="Codex 订阅账号与额度界面：模型选择、双周期额度与提醒；使用演示数据"></p>
 
+## 2.3.0 更新亮点
+
+- **官方桌面端可用**：DeepSeek Harness 官方桌面版（0.2.0-rc.2）已完成实机验收——安装插件、登录、模型列表、隐藏模型、带工具调用的对话、图片生成均通过；安装方式见[安装](#安装)。
+- **模型只来自你的账号**：模型列表完全读取账号目录，登录前或读取失败时不显示任何模型，也没有内置兜底；“模型与运行”里可逐个显示 / 隐藏输入框中的模型。
+- **设置页重新整理**：模型与搜索分成两个卡片，开关统一使用 DSH 自带控件，未登录时不再出现无效按钮。
+- **极速档 Ultrafast（Beta）**：OpenAI 发布会新增的速度档。只有账号模型目录明确给出该档位（目前为 Pro 500 的 GPT-6 Astra）才会在模型菜单里出现，其他账号不显示，也不会被悄悄降级到别的档位；由于暂无 Pro 500 账号做实机验证，标为 Beta。
+- **更短的支持诊断**：“复制诊断”现在只复制结论、近期失败和各能力状态（约 1 KB），完整报告仍可在详情里单独复制。
+- **子任务与稳定性**：Codex 子任务在 0.2.0 上真正生效，并且不再暴露账号里连接的应用；修复了图片生成在桌面版上被宿主拒绝的问题（2.2.11）。
+
 ## 功能一览
 
 把订阅接入、模型控制、额度管理和图片创作放在同一个 DSH 工作流里。下面逐项列出插件提供的能力；模型权限和额度以账号实际返回为准。
@@ -34,7 +43,7 @@
 | **模型目录同步** | 自动读取账号可用模型与推理档位，支持手动刷新；目录标注退役的模型会在模型名称旁显示退役日期，不会自动切换你的选择 | 新型号以账号目录开放情况为准；模型只来自账号目录，未登录或读取失败时不显示任何模型，也不使用内置列表 |
 | **模型列表** | 在“模型与运行”中逐个显示或隐藏输入框中的模型 | 仅影响列表显示；已选模型和请求不受影响 |
 | **推理等级** | 在输入框选择模型支持的推理档位 | 档位随模型变化 |
-| **高速模式** | 在支持的模型上切换标准 / Fast，输入框显示闪电标识 | 默认标准；Fast 会增加用量消耗 |
+| **速度档位** | 在支持的模型上切换标准 / Fast；账号目录提供时还有 Ultrafast（Beta），输入框显示闪电标识 | 默认标准；Fast 和 Ultrafast 会增加用量消耗，Ultrafast 目前仅 Pro 500 |
 | **输出详略** | 选择简短、适中或详细的回答偏好 | 默认跟随模型 |
 | **输入图片精度** | 为发送给聊天模型的图片选择低、高、原图或自动精度 | 影响输入 token 用量；默认自动 |
 | **订阅联网搜索** | 使用 Codex 搜索，或选择 DSH 搜索 | 可选自动、DSH、Codex 路由 |
@@ -81,7 +90,7 @@
 | **子任务组件管理** | 在设置里安装、停用或卸载官方 Codex 子任务组件 | 由支持的 DSH 宿主管理，无需额外登录 |
 | **云端上下文压缩** | 使用 Codex 云端压缩继续长对话，同时保留 DSH 历史与原生压缩 | Beta / 实验性；默认关闭，开启的请求使用 SSE |
 | **WebSocket 连接** | 实验性复用连接和上下文传输，连接失败可回退 SSE | Beta / 实验性；默认 SSE，不保证更快 |
-| **支持诊断** | 查看功能前置条件、近期操作结果与宿主注册信息；报告不含凭据和会话正文 | 设置 → 维护；未验证不代表故障 |
+| **支持诊断** | 查看功能前置条件和近期操作结果，一键复制约 1 KB 的精简报告（完整报告可另行复制）；不含凭据和会话正文 | 设置 → 维护；未验证不代表故障 |
 | **预测缓存清理** | 查看并清理本地额度预测历史 | 保留登录、草稿、原图和共享依赖 |
 
 [详细用法、限制与可选组件 →](docs/GUIDE.zh-CN.md)
@@ -101,16 +110,16 @@
 ### 在插件页面安装（推荐）
 
 1. 打开 DSH 的 **插件 → 添加插件**。
-2. 在 **包名或地址** 输入框中粘贴下面的包名：
+2. 在 **包名或地址** 输入框中粘贴下面的包名（已带版本号）：
 
    ```text
-   dsh-codex-subscription
+   dsh-codex-subscription@2.3.0
    ```
 
 3. 点击 **安装**，等待安装完成；按页面提示操作，需要重启时先保存工作。
 4. 打开 **设置 → Codex 订阅**，登录 ChatGPT，然后在会话中选择 Codex 模型。
 
-默认安装最新正式版。指定版本时填 `dsh-codex-subscription@版本号`，版本号见[发布页](https://github.com/WSL043/dsh-codex-subscription/releases)。
+为什么要带版本号：DSH 的插件页只会安装发布满 24 小时的版本，只填包名时，新版本发布后的第一天可能装到更早的旧版，而旧版不一定兼容你的 DSH。带上版本号可以直接装到这一版；发布一天后只填 `dsh-codex-subscription` 也会装到最新正式版。其他版本号见[发布页](https://github.com/WSL043/dsh-codex-subscription/releases)。
 
 <details>
 <summary>终端安装（已能运行 dsh 命令）</summary>
@@ -139,7 +148,7 @@ dsh --profile headless "只回复：ok"
 
 ### 日常操作，留在输入框
 
-选模型、调推理档位、开启高速模式，同时查看剩余额度。无需为每次请求打开设置。
+选模型、调推理档位、切换速度档位，同时查看剩余额度。无需为每次请求打开设置。
 
 <p align="center"><img src="docs/assets/composer-astra-fast.png" width="800" alt="DSH 输入框：GPT-6-Astra Max、闪电标识的 Fast 高速模式与剩余额度"></p>
 
@@ -273,9 +282,9 @@ dsh plugin --profile web remove dsh-codex-subscription
 ## 常见问题
 
 <details>
-<summary>DSH 0.2.0-rc.2 提示插件 2.2.4 不兼容？</summary>
+<summary>DSH 0.2.0-rc.2 提示插件不兼容？</summary>
 
-2.2.4 及更早版本发布时 DSH 0.2.0-rc.2 还不存在，其兼容声明无法事后修改。请更新到 2.2.5 或更高版本：在 **插件** 页面更新，或运行 `dsh plugin --profile web update dsh-codex-subscription`，然后重启 DSH。
+2.2.10 之前的所有版本发布时 DSH 0.2.0-rc.2 还不存在，其兼容声明无法事后修改。请安装 2.2.10 或更高版本（只填包名时，发布满 24 小时前可能装到旧版，请直接填 `dsh-codex-subscription@2.2.10`）：在 **插件** 页面更新，或运行 `dsh plugin --profile web update dsh-codex-subscription`，然后重启 DSH。
 
 </details>
 

package/lib/client.js

@@ -4977,6 +4977,8 @@ window.__ModuleLoader__.load({
 			quotaAlertsHint: "低于所选剩余阈值时提醒；自定义可分别设置短周期与长周期。仅使用新鲜数据，不弹窗打断。",
 			quotaWarning: "{window}额度剩余 {value}%，请留意用量与重置时间。",
 			speedFastAstraHint: "Astra 约 2 倍速度，消耗更多 Credits",
+			speedUltrafast: "极速 Beta",
+			speedUltrafastHint: "Ultrafast 档，仅账号目录提供时显示（目前为 Pro 500）；速度最快，额度和 Credits 消耗更快",
 			imageEditLocation: "位置",
 			imageEditReferenceGuide: "本次编辑的干净源图为「{sourceName}」，编号定位参考图为「{referenceName}」。坐标以图片左上角为原点，x 向右、y 向下，百分比相对于整张图片。请查看这两张图片，将它们同时作为编辑工具的参考图，并在工具提示词中完整保留下方编号、位置和修改要求。只修改源图中对应位置的内容；定位参考图上的编号、圆点和引线仅用于定位，不得绘入最终结果。若无法读取两张图片或确定位置，请说明问题，不要猜测或忽略标注。",
 			nav: "Codex 订阅",
@@ -5010,7 +5012,8 @@ window.__ModuleLoader__.load({
 			diagnostics: "支持诊断",
 			diagnosticsLoad: "生成诊断",
 			diagnosticsLoading: "生成中…",
-			diagnosticsCopy: "复制诊断",
+			diagnosticsCopy: "复制诊断（精简）",
+			diagnosticsCopyFull: "复制完整诊断",
 			diagnosticsCopied: "已复制",
 			diagnosticsFailed: "服务端诊断不可用，已生成可复制的本地脱敏诊断。请检查 DSH 服务连接与插件版本。",
 			feedbackOpen: "反馈问题",
@@ -5532,6 +5535,8 @@ window.__ModuleLoader__.load({
 			quotaAlertsHint: "Warn at the selected remaining threshold. Custom sets short and long windows separately. Fresh data only; no interrupting popups.",
 			quotaWarning: "{window}: {value}% remaining. Check usage and reset time.",
 			speedFastAstraHint: "About 2x speed on Astra; higher Credits use",
+			speedUltrafast: "Ultrafast Beta",
+			speedUltrafastHint: "Ultrafast tier, shown only when your account catalog offers it (Pro 500 today); fastest, uses allowance and Credits faster",
 			imageEditLocation: "Location",
 			imageEditReferenceGuide: "The clean source for this edit is \"{sourceName}\"; \"{referenceName}\" is the numbered location reference. Coordinates start at the top-left, x increases rightward and y downward; percentages refer to the whole image. Inspect both images and pass both as references to the image-editing tool. Preserve every number, position and requested change below in the tool prompt. Edit the corresponding content in the clean source; numbers, dots and leader lines on the location reference are guidance only and must not appear in the final result. If either image cannot be read or a location is unclear, explain the problem instead of guessing or ignoring annotations.",
 			nav: "Codex",
@@ -5565,7 +5570,8 @@ window.__ModuleLoader__.load({
 			diagnostics: "Support diagnostics",
 			diagnosticsLoad: "Create report",
 			diagnosticsLoading: "Creating…",
-			diagnosticsCopy: "Copy report",
+			diagnosticsCopy: "Copy report (short)",
+			diagnosticsCopyFull: "Copy full report",
 			diagnosticsCopied: "Copied",
 			diagnosticsFailed: "Server diagnostics unavailable. A local, redacted report is available to copy. Check the DSH service and plugin version.",
 			feedbackOpen: "Report a problem",
@@ -7047,7 +7053,6 @@ window.__ModuleLoader__.load({
 		const DEFAULT_SEARCH_PROVIDER = SEARCH_PROVIDER_AUTO;
 		const SPEED_MODE_FIELD = "speedMode";
 		const SPEED_MODE_STANDARD = "standard";
-		const SPEED_MODE_FAST = "fast";
 		const DEFAULT_SPEED_MODE = SPEED_MODE_STANDARD;
 		const OUTPUT_VERBOSITY_FIELD = "outputVerbosity";
 		const OUTPUT_VERBOSITY_DEFAULT = "default";
@@ -7115,7 +7120,11 @@ window.__ModuleLoader__.load({
 		].includes(value) ? value : DEFAULT_OUTPUT_VERBOSITY;
 		const normalizeInputImageDetail = (value) => INPUT_IMAGE_DETAILS.includes(value) ? value : DEFAULT_INPUT_IMAGE_DETAIL;
 		const normalizeStreamIdleTimeoutMinutes = (value) => STREAM_IDLE_TIMEOUT_MINUTES.includes(value) ? value : 10;
-		const normalizeSpeedMode = (value) => ["standard", "fast"].includes(value) ? value : DEFAULT_SPEED_MODE;
+		const normalizeSpeedMode = (value) => [
+			"standard",
+			"fast",
+			"ultrafast"
+		].includes(value) ? value : DEFAULT_SPEED_MODE;
 		const normalizeContextMode = (value) => [
 			"standard",
 			"extended",
@@ -7155,6 +7164,7 @@ window.__ModuleLoader__.load({
 			let availableModels = [];
 			let verbosityModels = [];
 			let fastModels;
+			let ultrafastModels = [];
 			let catalogStatus;
 			let modelsLoading = false;
 			let modelError = false;
@@ -7199,6 +7209,7 @@ window.__ModuleLoader__.load({
 					disabledModels: normalizeDisabledModels(value?.[DISABLED_MODELS_FIELD]),
 					verbosityModels,
 					fastModels,
+					ultrafastModels,
 					catalogStatus,
 					modelsLoading,
 					modelError,
@@ -7226,6 +7237,7 @@ window.__ModuleLoader__.load({
 					availableModels = Array.isArray(value?.availableModels) ? value.availableModels : [];
 					verbosityModels = Array.isArray(value?.verbosityModels) ? value.verbosityModels : [];
 					fastModels = Array.isArray(value?.fastModels) ? value.fastModels : void 0;
+					ultrafastModels = Array.isArray(value?.ultrafastModels) ? value.ultrafastModels : [];
 					catalogStatus = value?.catalogStatus;
 				}
 				fallbackStatus = "ready";
@@ -7270,6 +7282,8 @@ window.__ModuleLoader__.load({
 							availableModels = Array.isArray(value?.availableModels) ? value.availableModels : [];
 							verbosityModels = Array.isArray(value?.verbosityModels) ? value.verbosityModels : [];
 							fastModels = Array.isArray(value?.fastModels) ? value.fastModels : void 0;
+							ultrafastModels = Array.isArray(value?.ultrafastModels) ? value.ultrafastModels : [];
+							ultrafastModels = Array.isArray(value?.ultrafastModels) ? value.ultrafastModels : [];
 							catalogStatus = value?.catalogStatus;
 						}
 					} else acceptFallback(value);
@@ -7293,13 +7307,15 @@ window.__ModuleLoader__.load({
 					const nextAvailableModels = Array.isArray(value?.availableModels) ? value.availableModels : [];
 					const nextVerbosityModels = Array.isArray(value?.verbosityModels) ? value.verbosityModels : [];
 					const nextFastModels = Array.isArray(value?.fastModels) ? value.fastModels : void 0;
+					const nextUltrafastModels = Array.isArray(value?.ultrafastModels) ? value.ultrafastModels : [];
 					catalogStatus = value?.catalogStatus;
-					const changed = !sameModels(contextModels, nextContextModels) || !sameModels(availableModels, nextAvailableModels) || !sameModels(verbosityModels, nextVerbosityModels) || JSON.stringify(fastModels) !== JSON.stringify(nextFastModels);
+					const changed = !sameModels(contextModels, nextContextModels) || !sameModels(availableModels, nextAvailableModels) || !sameModels(verbosityModels, nextVerbosityModels) || JSON.stringify(fastModels) !== JSON.stringify(nextFastModels) || JSON.stringify(ultrafastModels) !== JSON.stringify(nextUltrafastModels);
 					if (changed) {
 						contextModels = nextContextModels;
 						availableModels = nextAvailableModels;
 						verbosityModels = nextVerbosityModels;
 						fastModels = nextFastModels;
+						ultrafastModels = nextUltrafastModels;
 						publish();
 					}
 					return changed;
@@ -7660,7 +7676,7 @@ window.__ModuleLoader__.load({
 		}
 		//#endregion
 		//#region src/version.js
-		const PACKAGE_VERSION = "2.2.10";
+		const PACKAGE_VERSION = "2.3.0";
 		//#endregion
 		//#region src/client-recovery.js
 		async function recoveryCall(rpc, endpoint, payload = {}, timeoutMs = 1e4) {
@@ -7950,7 +7966,10 @@ window.__ModuleLoader__.load({
 			const modelLabel = currentChoice?.model.name ?? t("selectModel");
 			const speedSupported = state.current?.provider === "openai-codex" && (preferenceSnapshot.fastModels?.includes(state.current?.model) ?? supportsCodexFastMode(state.current?.model));
 			const speedWritable = preferenceSnapshot.status === "ready" && preferenceSnapshot.writable === true;
-			const fast = speedSupported && preferenceSnapshot.speedMode === "fast";
+			const ultrafastSupported = state.current?.provider === "openai-codex" && (preferenceSnapshot.ultrafastModels ?? []).includes(state.current?.model);
+			const ultrafast = ultrafastSupported && preferenceSnapshot.speedMode === "ultrafast";
+			const fast = ultrafast || speedSupported && preferenceSnapshot.speedMode === "fast";
+			const speedKey = ultrafast ? "speedUltrafast" : fast ? "speedFast" : "speedStandard";
 			const verbositySupported = state.current?.provider === "openai-codex" && preferenceSnapshot.verbosityModels.includes(state.current?.model);
 			const verbosityWritable = preferenceSnapshot.status === "ready" && preferenceSnapshot.writable === true;
 			const verbosityItems = [
@@ -8158,21 +8177,32 @@ window.__ModuleLoader__.load({
 				className: "codexModelSelectSubmenu",
 				role: "menu",
 				"aria-label": t("speedTitle"),
-				children: [option({
-					key: SPEED_MODE_STANDARD,
-					label: t("speedStandard"),
-					description: t("speedStandardHint"),
-					selected: !fast,
-					disabled: !speedWritable,
-					onClick: () => chooseSpeed(SPEED_MODE_STANDARD)
-				}), option({
-					key: SPEED_MODE_FAST,
-					label: t("speedFast"),
-					description: t(state.current?.model === "gpt-6-astra" ? "speedFastAstraHint" : "speedFastHint"),
-					selected: fast,
-					disabled: !speedWritable,
-					onClick: () => chooseSpeed(SPEED_MODE_FAST)
-				})]
+				children: [
+					option({
+						key: SPEED_MODE_STANDARD,
+						label: t("speedStandard"),
+						description: t("speedStandardHint"),
+						selected: !fast,
+						disabled: !speedWritable,
+						onClick: () => chooseSpeed(SPEED_MODE_STANDARD)
+					}),
+					speedSupported && option({
+						key: "fast",
+						label: t("speedFast"),
+						description: t(state.current?.model === "gpt-6-astra" ? "speedFastAstraHint" : "speedFastHint"),
+						selected: fast && !ultrafast,
+						disabled: !speedWritable,
+						onClick: () => chooseSpeed("fast")
+					}),
+					ultrafastSupported && option({
+						key: "ultrafast",
+						label: t("speedUltrafast"),
+						description: t("speedUltrafastHint"),
+						selected: ultrafast,
+						disabled: !speedWritable,
+						onClick: () => chooseSpeed("ultrafast")
+					})
+				]
 			});
 			else if (pane === "verbosity") submenu = /* @__PURE__ */ (0, react_jsx_runtime.jsx)("div", {
 				className: "codexModelSelectSubmenu",
@@ -8231,7 +8261,7 @@ window.__ModuleLoader__.load({
 					children: [
 						cell("model", t("modelLabel"), modelLabel),
 						reasoning === void 0 ? null : cell("effort", t("effortLabel"), effortLabel),
-						speedSupported && cell("speed", t("speedTitle"), t(fast ? "speedFast" : "speedStandard")),
+						(speedSupported || ultrafastSupported) && cell("speed", t("speedTitle"), t(speedKey)),
 						verbositySupported && cell("verbosity", t("verbosityTitle"), verbosityLabel),
 						submenu
 					]
@@ -9919,6 +9949,38 @@ window.__ModuleLoader__.load({
 			};
 		}
 		//#endregion
+		//#region src/diagnostic-compact.js
+		function compactDiagnostic(report) {
+			const summary = report.summary ?? { findings: [] };
+			return {
+				format: "compact-1",
+				plugin: report.version,
+				dsh: report.client?.version,
+				generatedAt: report.generatedAt,
+				account: report.account?.status,
+				catalog: report.catalog ? {
+					source: report.catalog.source,
+					refresh: report.catalog.refresh
+				} : void 0,
+				configuration: report.configuration,
+				findings: summary.findings.map((item) => ({
+					code: item.code,
+					severity: item.severity,
+					...item.code === "recent-failure" ? { failures: item.evidence.map((event) => ({
+						action: event.action,
+						source: event.source,
+						at: new Date(event.observedAt).toISOString()
+					})) } : {}
+				})),
+				capabilities: (report.inspection?.capabilities ?? []).filter((item) => item.readiness !== "not-applicable").map((item) => ({
+					id: item.id,
+					reason: item.reason,
+					...item.latest ? { latest: `${item.latest.action} ${item.latest.status}` } : { latest: "not-verified" }
+				})),
+				...report.client?.status === "unknown" ? { hostCollected: false } : {}
+			};
+		}
+		//#endregion
 		//#region src/client-diagnostics.jsx
 		function DiagnosticsCard({ rpc, t, diagnostics }) {
 			const [report, setReport] = (0, react.useState)();
@@ -9943,10 +10005,10 @@ window.__ModuleLoader__.load({
 					});
 				}).finally(() => setBusy(false));
 			};
-			const copy = () => {
+			const copy = (full = false) => {
 				if (report === void 0) return;
 				setCopyError(false);
-				Promise.resolve().then(() => navigator.clipboard.writeText(JSON.stringify(report, null, 2))).then(() => setCopied(true)).catch(() => setCopyError(true));
+				Promise.resolve().then(() => navigator.clipboard.writeText(JSON.stringify(full ? report : compactDiagnostic(report), null, 2))).then(() => setCopied(true)).catch(() => setCopyError(true));
 			};
 			return /* @__PURE__ */ (0, react_jsx_runtime.jsxs)("div", {
 				className: "codexSubscriptionCard codexSubscriptionDiagnostics",
@@ -9972,7 +10034,7 @@ window.__ModuleLoader__.load({
 								report === void 0 ? null : /* @__PURE__ */ (0, react_jsx_runtime.jsx)(Button, {
 									type: "button",
 									variant: "outline",
-									onClick: copy,
+									onClick: () => copy(),
 									children: copied ? t("diagnosticsCopied") : t("diagnosticsCopy")
 								}),
 								/* @__PURE__ */ (0, react_jsx_runtime.jsx)("a", {
@@ -10038,7 +10100,16 @@ window.__ModuleLoader__.load({
 							className: "codexSubscriptionHelp",
 							children: report.client?.topology?.length ? t("diagnosticsHostCollected") : t("diagnosticsHostUnknown")
 						}),
-						/* @__PURE__ */ (0, react_jsx_runtime.jsxs)("details", { children: [/* @__PURE__ */ (0, react_jsx_runtime.jsx)("summary", { children: t("diagnosticsDetails") }), /* @__PURE__ */ (0, react_jsx_runtime.jsx)("pre", { children: JSON.stringify(report, null, 2) })] })
+						/* @__PURE__ */ (0, react_jsx_runtime.jsxs)("details", { children: [
+							/* @__PURE__ */ (0, react_jsx_runtime.jsx)("summary", { children: t("diagnosticsDetails") }),
+							/* @__PURE__ */ (0, react_jsx_runtime.jsx)(Button, {
+								type: "button",
+								variant: "outline",
+								onClick: () => copy(true),
+								children: t("diagnosticsCopyFull")
+							}),
+							/* @__PURE__ */ (0, react_jsx_runtime.jsx)("pre", { children: JSON.stringify(report, null, 2) })
+						] })
 					] }),
 					error ? /* @__PURE__ */ (0, react_jsx_runtime.jsx)("p", {
 						className: "codexSubscriptionError",

package/lib/index.js

@@ -206,6 +206,7 @@ const DEFAULT_SEARCH_PROVIDER = SEARCH_PROVIDER_AUTO;
 const SPEED_MODE_FIELD = "speedMode";
 const SPEED_MODE_STANDARD = "standard";
 const SPEED_MODE_FAST = "fast";
+const SPEED_MODE_ULTRAFAST = "ultrafast";
 const DEFAULT_SPEED_MODE = SPEED_MODE_STANDARD;
 const OUTPUT_VERBOSITY_FIELD = "outputVerbosity";
 const OUTPUT_VERBOSITY_DEFAULT = "default";
@@ -364,7 +365,11 @@ const PREFERENCE_FIELDS = Object.freeze({
 		error: "Invalid search provider preference"
 	},
 	[SPEED_MODE_FIELD]: {
-		choices: [SPEED_MODE_STANDARD, SPEED_MODE_FAST],
+		choices: [
+			SPEED_MODE_STANDARD,
+			SPEED_MODE_FAST,
+			SPEED_MODE_ULTRAFAST
+		],
 		default: DEFAULT_SPEED_MODE,
 		error: "Invalid speed mode preference"
 	},
@@ -2878,6 +2883,7 @@ function normalizeTransportEvent(event, signal) {
 //#endregion
 //#region src/pi-ai-runtime.js
 const FAST_SERVICE_TIER = "priority";
+const ULTRAFAST_SERVICE_TIER = "ultrafast";
 /** Apply the chosen detail to user and tool-result images after the provider assembles the request. */
 function withInputImageDetail(payload, detail) {
 	if (!Array.isArray(payload.input)) return payload;
@@ -2937,13 +2943,14 @@ function openaiCodexSubscriptionProvider({ resolveSpeedMode = () => void 0, reso
 		const metadata = modelMetadata(model);
 		const requestedVerbosity = resolveOutputVerbosity();
 		const textVerbosity = supportsVerbosity(model) ? requestedVerbosity === "default" ? metadata?.defaultVerbosity ?? "medium" : requestedVerbosity : void 0;
-		const fast = resolveSpeedMode() === "fast" && (metadata?.supportsFast ?? supportsCodexFastMode(model?.id));
+		const speedMode = resolveSpeedMode();
+		const serviceTier = speedMode === "ultrafast" && metadata?.supportsUltrafast === true ? ULTRAFAST_SERVICE_TIER : speedMode === "fast" && (metadata?.supportsFast ?? supportsCodexFastMode(model?.id)) ? FAST_SERVICE_TIER : void 0;
 		const inputImageDetail = normalizeInputImageDetail(resolveInputImageDetail());
 		const onPayload = options.onPayload;
 		return {
 			...options,
 			...textVerbosity === void 0 ? {} : { textVerbosity },
-			...fast ? { serviceTier: FAST_SERVICE_TIER } : {},
+			...serviceTier === void 0 ? {} : { serviceTier },
 			async onPayload(payload, requestModel) {
 				const preferred = {
 					...payload,
@@ -2951,7 +2958,7 @@ function openaiCodexSubscriptionProvider({ resolveSpeedMode = () => void 0, reso
 						...payload.text ?? {},
 						verbosity: textVerbosity
 					} },
-					...fast ? { service_tier: FAST_SERVICE_TIER } : {}
+					...serviceTier === void 0 ? {} : { service_tier: serviceTier }
 				};
 				const managed = compaction?.preparePayload(preferred, model?.contextWindow) ?? preferred;
 				const next = await onPayload?.(managed, requestModel);
@@ -2962,7 +2969,7 @@ function openaiCodexSubscriptionProvider({ resolveSpeedMode = () => void 0, reso
 						...detailed.text ?? {},
 						verbosity: textVerbosity
 					} },
-					...fast ? { service_tier: FAST_SERVICE_TIER } : {}
+					...serviceTier === void 0 ? {} : { service_tier: serviceTier }
 				};
 			}
 		};
@@ -3051,7 +3058,7 @@ async function readSubscriptionCredentials(getAuth, readCredential, signal) {
 }
 //#endregion
 //#region src/version.js
-const PACKAGE_VERSION = "2.2.10";
+const PACKAGE_VERSION = "2.3.0";
 const USER_AGENT = `dsh-codex-subscription/${PACKAGE_VERSION}`;
 //#endregion
 //#region src/model-catalog.js
@@ -3114,6 +3121,7 @@ function reasoningMap(levels) {
 	return map;
 }
 const capabilityNames = (values) => [...new Set(values.filter((value) => typeof value === "string" && /^[a-z][a-z0-9_-]{0,31}$/u.test(value)))].sort().slice(0, 16);
+const speedTierIds = (value) => [...Array.isArray(value.additional_speed_tiers) ? value.additional_speed_tiers : [], ...Array.isArray(value.service_tiers) ? value.service_tiers.map((tier) => tier?.id) : []];
 function unsupportedCapabilities(value) {
 	const reasoning = capabilityNames((value.supported_reasoning_levels ?? []).map((item) => item?.effort)).filter((level) => !["none", ...LEVELS.slice(1)].includes(level));
 	const inputs = capabilityNames(Array.isArray(value.input_modalities) ? value.input_modalities : []).filter((input) => !["text", "image"].includes(input));
@@ -3122,7 +3130,8 @@ function unsupportedCapabilities(value) {
 		"default",
 		"standard",
 		"fast",
-		"priority"
+		"priority",
+		"ultrafast"
 	].includes(tier));
 	return {
 		...reasoning.length ? { reasoning } : {},
@@ -3166,7 +3175,8 @@ function visibleModel(value, displayNames, now) {
 			"medium",
 			"high"
 		].includes(value.default_verbosity) ? value.default_verbosity : void 0,
-		supportsFast: [...Array.isArray(value.additional_speed_tiers) ? value.additional_speed_tiers : [], ...Array.isArray(value.service_tiers) ? value.service_tiers.map((tier) => tier?.id) : []].some((tier) => tier === "fast" || tier === "priority")
+		supportsFast: speedTierIds(value).some((tier) => tier === "fast" || tier === "priority"),
+		supportsUltrafast: speedTierIds(value).includes("ultrafast")
 	};
 }
 function parseOfficialModelCatalog(value, now = Date.now()) {
@@ -3981,10 +3991,7 @@ function createCodexImageTool(options) {
 			};
 			if (exec.parent !== void 0) exec.deferContext(createUserMessage({
 				content: imageContent(result),
-				source: {
-					kind: "plugin",
-					plugin: "codex-subscription"
-				}
+				source: { kind: "codex-subscription-image" }
 			}));
 			return result;
 		}
@@ -5832,6 +5839,7 @@ function createSubscriptionRpcHandler({ authHandler, usageReader, resetCreditSer
 					contextModels: Array.isArray(value?.contextModels) ? value.contextModels : [],
 					verbosityModels: Array.isArray(value?.verbosityModels) ? value.verbosityModels : [],
 					fastModels: Array.isArray(value?.fastModels) ? value.fastModels : [],
+					ultrafastModels: Array.isArray(value?.ultrafastModels) ? value.ultrafastModels : [],
 					catalogStatus: value?.catalogStatus
 				}
 			};
@@ -6103,6 +6111,7 @@ function apply(ctx, config = {}) {
 				})),
 				catalogStatus: modelCatalog.status(),
 				verbosityModels: provider.getModels().filter((model) => modelCatalog.metadata(model.id)?.supportVerbosity ?? model.id !== "gpt-5.3-codex-spark").map((model) => model.id),
+				ultrafastModels: provider.getModels().filter((model) => modelCatalog.metadata(model.id)?.supportsUltrafast === true).map((model) => model.id),
 				fastModels: provider.getModels().filter((model) => modelCatalog.metadata(model.id)?.supportsFast ?? supportsCodexFastMode(model.id)).map((model) => model.id),
 				writable: ctx.settings.writable
 			};

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "dsh-codex-subscription",
-  "version": "2.2.10",
+  "version": "2.3.0",
   "description": "Use ChatGPT and Codex subscriptions in DeepSeek Harness with OAuth, quota, safe resets, web search, images, and Fast mode",
   "type": "module",
   "main": "./lib/index.js",