dsh-client-auto-continue 0.9.0 → 0.10.1

package/README.md

@@ -8,7 +8,7 @@
 <h1 align="center">dsh-auto-continue</h1>
 
 <p align="center">
-  <em>DSH Web UI plugin — when a request is interrupted by a network error or any other non-human cause, it automatically types 「继续」 and sends it for you.</em>
+  <em>DSH Web UI plugin — when a request is interrupted by a network error or any other non-human cause, it automatically sends “Continue” for you.</em>
 </p>
 
 <p align="center">
@@ -17,6 +17,7 @@
   <a href="https://github.com/HsiangNianian/dsh-auto-continue/stargazers"><img src="https://img.shields.io/github/stars/HsiangNianian/dsh-auto-continue?logo=github&label=Stars" alt="GitHub stars"></a>
   <a href="https://github.com/HsiangNianian/dsh-auto-continue/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-MIT-65a30d?style=flat" alt="MIT license"></a>
   <a href="https://awesome-dsh-plugin.com"><img src="https://awesome-dsh-plugin.com/badge.svg" alt="awesome · DSH plugin"></a>
+  <a href="https://www.dsh.so/artifact/dsh-auto-continue/"><img src="https://www.dsh.so/badge/install/dsh-auto-continue.svg" alt="dsh.so install"></a>
   <br>
   <img src="https://img.shields.io/badge/TypeScript-3178C6?style=flat&logo=typescript&logoColor=fff" alt="TypeScript">
   <img src="https://img.shields.io/badge/esbuild-FFCF00?style=flat&logo=esbuild&logoColor=000" alt="esbuild">
@@ -31,7 +32,7 @@
 
 ## What It Does
 
-For [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) (`dsh web`): whenever a request in the web GUI gets interrupted by a **non-human cause**, the plugin simulates the user typing **「继续」** and sends it, so the agent keeps working without manual intervention. The message enters the session log exactly like a manual prompt — the model sees it, and the interrupted work resumes. Since 0.8.0 the engine runs **inside the host process** (single instance), so it keeps watching even with every browser tab closed, and multiple open tabs can never double-send.
+For [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) (`dsh web`): whenever a request in the web GUI gets interrupted by a **non-human cause**, the plugin simulates the user typing **“Continue”** and sends it, so the agent keeps working without manual intervention. The message enters the session log exactly like a manual prompt — the model sees it, and the interrupted work resumes. Since 0.8.0 the engine runs **inside the host process** (single instance), so it keeps watching even with every browser tab closed, and multiple open tabs can never double-send.
 
 ![demo](docs/demo.svg)
 
@@ -39,7 +40,8 @@ For [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) (`dsh we
 
 - **Error classification** — transient failures (network / timeout / 5xx / 429…) are auto-resumed; permanent ones are **skipped** and notified, because retrying them never helps. A failure counts as permanent when its HTTP status is 401/403 or its code/message matches auth, credential/API-key, balance/quota, unknown-model, or context-length/overflow keywords. Provider-specific exceptions can be opted into with literal custom retryable patterns; turn classification off to resume everything
 - **Adaptive backoff** — consecutive failures wait longer each time (cooldown × factor: 20s → 40s → 80s…), capped at the max backoff, instead of hammering a broken upstream
-- **Templated continue text** — `continueText` supports `{code}` `{message}` `{status}` `{tool}` `{turn}` `{errorCount}` `{sessionTitle}` `{elapsed}` placeholders, so the resume message can carry the failure context ("继续 (git push failed: UPSTREAM)"); a **separate template** fires on `max-tokens` (e.g. "继续输出, 不要重复已生成的内容")
+- **English / Chinese localization** — the settings card, built-in resume / guard / loop text, and browser notifications follow DSH's active UI language (initially selected from the browser language). Only `en` and `zh` are supported; other languages fall back to Chinese. Switching languages updates built-in defaults without overwriting custom text
+- **Templated continue text** — `continueText` supports `{code}` `{message}` `{status}` `{tool}` `{turn}` `{errorCount}` `{sessionTitle}` `{elapsed}` placeholders, so the resume message can carry the failure context ("Continue ({tool} failed: {code})"); a **separate template** fires on `max-tokens` (e.g. "Continue the output without repeating anything already generated")
 - **Idempotency guard** — before resuming, the plugin inspects the last tool call: if its result is unconfirmed (the turn died mid-tool, e.g. a `git push` that may have gone through), the resume message tells the model to check state first and not to rerun; if the tool is confirmed done, it says so and asks not to repeat it; a failed tool gets no guard (retrying it is the point). Both guard texts are configurable (`{tool}` / `{result}` placeholders)
 - **Pause** — a global **Pause auto-continue** toggle in the settings card stops everything (live + scan) instantly; per-session pauses (e.g. via a notification button) suspend only one session until they expire. The **Resume now** notification button is the one explicit exception: pressing it is the user asking for exactly one send, pause or not
 - **Notification buttons** — notifications carry **Resume now** (send immediately, ignoring cooldown, the consecutive cap and any pause) and **Pause this session 1h** actions
@@ -146,16 +148,19 @@ dsh web
 
 Everything is configurable from the GUI — no file or console edits needed. Open **Settings → Plugins** and find the **dsh-client-auto-continue** configuration card, right where every other plugin's config lives. Besides the fields below, the card shows a live **stats panel** (today's activity with a reset button) and the list of **paused sessions** (each with a per-session resume button).
 
-**Or skip the GUI and edit the config file directly** — the engine reads the plugin's section from `~/.dsh/settings.yaml` (one shared file for every plugin's sections), so this works in any install, patched or not. The file is watched and re-read automatically, so changes apply live; restart `dsh web` if a page that was already open doesn't pick them up. Fields you leave out fall back to the defaults in the table below:
+**Or skip the GUI and edit the config file directly** — the engine reads the plugin's section from `~/.dsh/settings.yaml` (one shared file for every plugin's sections), so this works in any install, patched or not. The file is watched and re-read automatically, so changes apply live; restart `dsh web` if a page that was already open doesn't pick them up. Fields you leave out fall back to the defaults in the table below.
+
+The browser mirrors DSH's active language into the internal `locale` field. Leave the five localized text fields empty or omit them to follow that language automatically; any non-empty value is treated as your own template and is never rewritten when the language changes:
 
 ```yaml
 auto-continue:
+  locale: 'en' # normally managed by the browser
   paused: false
-  continueText: '继续'
-  continueTextMaxTokens: '继续'
+  continueText: ''
+  continueTextMaxTokens: ''
   guardTools: true
-  guardPendingText: '(上一步工具「{tool}」可能未完成, 先确认状态再继续, 不要重复执行)'
-  guardDoneText: '(上一步工具「{tool}」已完成, 结果: {result}; 不要重复执行, 直接继续)'
+  guardPendingText: ''
+  guardDoneText: ''
   graceMs: 3000
   cooldownMs: 20000
   maxConsecutive: 3
@@ -174,7 +179,7 @@ auto-continue:
   loopShortCount: 12
   loopRepeatText: 4
   loopToolRepeat: 5
-  loopText: '(检测到你可能陷入循环, 请停止重复刚才的动作, 换一种方式继续)'
+  loopText: ''
 ```
 
 **How the card works:**
@@ -191,8 +196,8 @@ auto-continue:
 | Field | Default | Description |
 | --- | --- | --- |
 | Pause auto-continue | `off` | Global pause: no live or scan auto-send fires, queued pending sends are cancelled |
-| Continue text | `继续` | Text automatically sent after an interruption |
-| Continue text (max tokens) | `继续` | Text sent when the output token ceiling is reached (same placeholders) |
+| Continue text | `Continue` | Text automatically sent after an interruption |
+| Continue text (max tokens) | `Continue` | Text sent when the output token ceiling is reached (same placeholders) |
 | Idempotency guard | `on` | Inspect the last tool call before resuming and steer the model (see What It Does) |
 | Loop guard | `on` | Detect a running turn spinning in place and restart it (see What It Does) |
 | Short-sentence max (chars) | `40` | A model message shorter than this counts as a short sentence (spinning signal) |
@@ -200,9 +205,9 @@ auto-continue:
 | Short-sentence threshold | `12` | Consecutive short sentences inside the window, with no tool call in between, trip the loop guard |
 | Identical message count | `4` | Consecutive identical messages (any length) trip the loop guard — the strongest spinning signal |
 | Same-tool repeat count | `5` | Consecutive calls of the same tool with identical arguments and results trip the loop guard |
-| Loop text | `(检测到你可能陷入循环, 请停止重复刚才的动作, 换一种方式继续)` | Text sent after the loop guard restarts a turn; `{tool}` placeholder |
-| Guard text (unconfirmed result) | `(上一步工具「{tool}」可能未完成, 先确认状态再继续, 不要重复执行)` | Appended when the last tool may have partially executed; `{tool}` placeholder |
-| Guard text (tool succeeded) | `(上一步工具「{tool}」已完成, 结果: {result}; 不要重复执行, 直接继续)` | Appended when the last tool is confirmed done; `{tool}` / `{result}` placeholders |
+| Loop text | `(You may be stuck in a loop. Stop repeating the last action and continue with a different approach.)` | Text sent after the loop guard restarts a turn; `{tool}` placeholder |
+| Guard text (unconfirmed result) | `(The previous tool "{tool}" may not have completed. Check its state before continuing and do not run it again.)` | Appended when the last tool may have partially executed; `{tool}` placeholder |
+| Guard text (tool succeeded) | `(The previous tool "{tool}" completed successfully. Result: {result}; do not run it again. Continue from there.)` | Appended when the last tool is confirmed done; `{tool}` / `{result}` placeholders |
 | Grace period (ms) | `3000` | Wait after an interruption; cancelled if the host recovers on its own |
 | Cooldown (ms) | `20000` | Min interval between auto-continues per session (failed attempts count too) |
 | Max consecutive | `3` | Max consecutive auto-continues; stops until a user intervenes or a turn completes |
@@ -226,7 +231,7 @@ auto-continue:
 
 Patterns are literal substrings, not regular expressions. Blank lines are ignored; any matching line wins before the built-in permanent-error rules. Cooldown and consecutive-attempt limits still apply.
 
-`continueText` (and `continueTextMaxTokens`) accept the placeholders `{code}`, `{message}`, `{status}`, `{tool}` (last tool call before the failure), `{turn}`, `{errorCount}` (consecutive failures including this one), `{sessionTitle}` (from the session list) and `{elapsed}` (time since the failure, e.g. `1m5s`) — e.g. `继续 ({tool}: {code})` becomes `继续 (git push: UPSTREAM)`. The guard texts accept `{tool}` and `{result}` (a truncated excerpt of the last tool output).
+`continueText` (and `continueTextMaxTokens`) accept the placeholders `{code}`, `{message}`, `{status}`, `{tool}` (last tool call before the failure), `{turn}`, `{errorCount}` (consecutive failures including this one), `{sessionTitle}` (from the session list) and `{elapsed}` (time since the failure, e.g. `1m5s`) — e.g. `Continue ({tool}: {code})` becomes `Continue (git push: UPSTREAM)`. The guard texts accept `{tool}` and `{result}` (a truncated excerpt of the last tool output).
 
 ---
 

package/README.zh.md

@@ -17,6 +17,7 @@
   <a href="https://github.com/HsiangNianian/dsh-auto-continue/stargazers"><img src="https://img.shields.io/github/stars/HsiangNianian/dsh-auto-continue?logo=github&label=Stars" alt="GitHub stars"></a>
   <a href="https://github.com/HsiangNianian/dsh-auto-continue/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-MIT-65a30d?style=flat" alt="MIT license"></a>
   <a href="https://awesome-dsh-plugin.com"><img src="https://awesome-dsh-plugin.com/badge.svg" alt="awesome · DSH plugin"></a>
+  <a href="https://www.dsh.so/artifact/dsh-auto-continue/"><img src="https://www.dsh.so/badge/install/dsh-auto-continue.svg" alt="dsh.so install"></a>
   <br>
   <img src="https://img.shields.io/badge/TypeScript-3178C6?style=flat&logo=typescript&logoColor=fff" alt="TypeScript">
   <img src="https://img.shields.io/badge/esbuild-FFCF00?style=flat&logo=esbuild&logoColor=000" alt="esbuild">
@@ -39,6 +40,7 @@
 
 - **错误分类** — 临时性错误(网络 / 超时 / 5xx / 429 等)自动续跑; 永久性错误跳过并通知, 因为重试也没用。判定为永久性的条件: HTTP 状态码 401/403, 或 code/message 命中认证、凭据/API Key、余额/配额、模型不存在、上下文长度/超限等关键词。provider 专属例外可由用户显式填写普通文本匹配; 关闭分类后则全部自动继续
 - **自适应退避** — 连续失败时等待时间递增(冷却 × 系数: 20s → 40s → 80s…), 有上限, 不再对故障上游狂轰滥炸
+- **中英文本地化** — 设置卡片、内置续跑 / 护栏 / 循环文案以及浏览器通知都跟随 DSH 当前界面语言(初始值来自浏览器语言)。只支持 `en` 和 `zh`, 其他语言回落中文; 切换语言只会替换内置默认值, 不会覆盖用户自定义文本
 - **模板化继续文本** — `continueText` 支持 `{code}` `{message}` `{status}` `{tool}` `{turn}` `{errorCount}` `{sessionTitle}` `{elapsed}` 占位符, 续跑消息可携带失败上下文(如「继续 (git push 失败: UPSTREAM)」); 达到 `max-tokens` 时使用**另一套模板**(如「继续输出, 不要重复已生成的内容」)
 - **幂等护栏** — 续跑前检查上一步工具调用: 结果未确认(回合在工具执行中途夭折, 如 `git push` 可能已经推上去了)时, 续跑消息会提示模型先确认状态、不要重复执行; 工具已确认成功时说明已完成、请勿重复; 工具失败则不加护栏(重试本来就是目的)。两段护栏文本都可配置(支持 `{tool}` / `{result}` 占位符)
 - **暂停** — 设置卡片里的全局 **暂停自动继续** 开关可立即停掉一切(实时 + 扫描); 会话级暂停(如通过通知按钮)只挂起单个会话, 到期自动恢复。唯一的显式例外是通知里的 **立即续跑** 按钮——按下它等于用户明确要求发送这一次, 不受暂停限制
@@ -146,16 +148,19 @@ dsh web
 
 所有参数都可以在 GUI 里配置——无需改文件或控制台。打开 **设置 → 插件**, 找到 **dsh-client-auto-continue** 的配置卡片, 和其他插件的配置放在一起。除了下面的字段, 卡片还带一个**统计面板**(今日活动, 可一键清零)和**已暂停会话**列表(每个都可单独解除)。
 
-**或者跳过 GUI, 直接编辑配置文件** — 引擎从 `~/.dsh/settings.yaml` 读取插件段落(所有插件的设置都在这一个文件里), 因此无论是否打过补丁都能用。文件被监听、自动重读, 改动即时生效; 若已打开的页面没反应, 重启 `dsh web` 即可。没填写的字段会回落到下表中的默认值:
+**或者跳过 GUI, 直接编辑配置文件** — 引擎从 `~/.dsh/settings.yaml` 读取插件段落(所有插件的设置都在这一个文件里), 因此无论是否打过补丁都能用。文件被监听、自动重读, 改动即时生效; 若已打开的页面没反应, 重启 `dsh web` 即可。没填写的字段会回落到下表中的默认值。
+
+浏览器会把 DSH 当前语言同步到内部 `locale` 字段。下面五个本地化文本字段保持留空或直接省略时, 会自动跟随语言; 任何非空值都视为用户自己的模板, 切换语言时不会改写:
 
 ```yaml
 auto-continue:
+  locale: 'zh' # 通常由浏览器自动维护
   paused: false
-  continueText: '继续'
-  continueTextMaxTokens: '继续'
+  continueText: ''
+  continueTextMaxTokens: ''
   guardTools: true
-  guardPendingText: '(上一步工具「{tool}」可能未完成, 先确认状态再继续, 不要重复执行)'
-  guardDoneText: '(上一步工具「{tool}」已完成, 结果: {result}; 不要重复执行, 直接继续)'
+  guardPendingText: ''
+  guardDoneText: ''
   graceMs: 3000
   cooldownMs: 20000
   maxConsecutive: 3
@@ -174,7 +179,7 @@ auto-continue:
   loopShortCount: 12
   loopRepeatText: 4
   loopToolRepeat: 5
-  loopText: '(检测到你可能陷入循环, 请停止重复刚才的动作, 换一种方式继续)'
+  loopText: ''
 ```
 
 **卡片操作说明:**

package/lib/client.js

@@ -35,6 +35,50 @@ __export(index_exports, {
 });
 module.exports = __toCommonJS(index_exports);
 
+// src/shared/core.ts
+var LOCALIZED_TEXT_DEFAULTS = {
+  zh: {
+    continueText: "继续",
+    continueTextMaxTokens: "继续",
+    guardPendingText: "(上一步工具「{tool}」可能未完成, 先确认状态再继续, 不要重复执行)",
+    guardDoneText: "(上一步工具「{tool}」已完成, 结果: {result}; 不要重复执行, 直接继续)",
+    loopText: "(检测到你可能陷入循环, 请停止重复刚才的动作, 换一种方式继续)"
+  },
+  en: {
+    continueText: "Continue",
+    continueTextMaxTokens: "Continue",
+    guardPendingText: '(The previous tool "{tool}" may not have completed. Check its state before continuing and do not run it again.)',
+    guardDoneText: '(The previous tool "{tool}" completed successfully. Result: {result}; do not run it again. Continue from there.)',
+    loopText: "(You may be stuck in a loop. Stop repeating the last action and continue with a different approach.)"
+  }
+};
+var DEFAULT_CONFIG = {
+  locale: "zh",
+  ...LOCALIZED_TEXT_DEFAULTS.zh,
+  guardTools: true,
+  graceMs: 3e3,
+  cooldownMs: 2e4,
+  maxConsecutive: 3,
+  scanOnBoot: true,
+  scanLimit: 8,
+  freshMs: 15 * 60 * 1e3,
+  verbose: true,
+  classify: true,
+  retryableErrorPatterns: "",
+  backoffFactor: 2,
+  backoffMaxMs: 3e5,
+  notify: false,
+  paused: false,
+  loopGuard: true,
+  loopShortChars: 40,
+  loopWindowMs: 3e4,
+  loopShortCount: 12,
+  loopRepeatText: 4,
+  loopToolRepeat: 5
+};
+var RECOVERY_WINDOW_MS = 10 * 60 * 1e3;
+var ECHO_WINDOW_MS = 10 * 60 * 1e3;
+
 // src/client/locales.ts
 var zh = {
   "card.title": "自动继续",
@@ -43,14 +87,18 @@ var zh = {
   "field.pausedHint": "全局暂停: 实时与扫描都不会再自动发送, 已排队的待发送也会取消。",
   "field.continueText": "继续文本",
   "field.continueTextHint": "中断后自动发送的消息内容。",
+  "default.continueText": LOCALIZED_TEXT_DEFAULTS.zh.continueText,
   "field.continueTextMaxTokens": "超限时的继续文本",
   "field.continueTextMaxTokensHint": "达到输出 token 上限时自动发送的文本, 支持与继续文本相同的占位符。",
+  "default.continueTextMaxTokens": LOCALIZED_TEXT_DEFAULTS.zh.continueTextMaxTokens,
   "field.guardTools": "幂等护栏",
   "field.guardToolsHint": "续跑前检查上一步工具调用: 结果未确认时提示先确认状态, 已成功时提示不要重复执行, 避免重复 commit/调 API。",
   "field.guardPendingText": "结果未确认时的护栏文本",
   "field.guardPendingTextHint": "上一步工具可能已部分执行时附加到继续文本之后, 支持 {tool} 占位符。",
+  "default.guardPendingText": LOCALIZED_TEXT_DEFAULTS.zh.guardPendingText,
   "field.guardDoneText": "工具已成功时的护栏文本",
   "field.guardDoneTextHint": "上一步工具已确认成功时附加到继续文本之后, 支持 {tool} 与 {result}(结果摘要)占位符。",
+  "default.guardDoneText": LOCALIZED_TEXT_DEFAULTS.zh.guardDoneText,
   "field.graceMs": "宽限期 (ms)",
   "field.graceMsHint": "检测到中断后等待的时长; 期间宿主自行恢复则取消。",
   "field.cooldownMs": "冷却时间 (ms)",
@@ -96,6 +144,7 @@ var zh = {
   "field.loopToolRepeatHint": "同工具+同参数+同结果的连续调用多少次时判定死循环; 参数或结果有变化视为有进展。",
   "field.loopText": "循环提示文本",
   "field.loopTextHint": "打断后重启回合时发送的文本, 支持 {tool} 占位符。",
+  "default.loopText": LOCALIZED_TEXT_DEFAULTS.zh.loopText,
   "stats.byCode": "按错误码统计",
   "stats.empty": "今天还没有自动继续记录。",
   "stats.reset": "清零",
@@ -121,19 +170,23 @@ var zh = {
 };
 var en = {
   "card.title": "Auto continue",
-  "card.description": "When a request is interrupted by a non-human cause, automatically send 「继续」 to resume.",
+  "card.description": 'When a request is interrupted by a non-human cause, automatically send "Continue" to resume.',
   "field.paused": "Pause auto-continue",
   "field.pausedHint": "Globally pause: no live or scan auto-send fires, and queued pending sends are cancelled.",
   "field.continueText": "Continue text",
   "field.continueTextHint": "Message automatically sent after an interruption.",
+  "default.continueText": LOCALIZED_TEXT_DEFAULTS.en.continueText,
   "field.continueTextMaxTokens": "Continue text (max tokens)",
   "field.continueTextMaxTokensHint": "Text sent when the output token ceiling is reached; same placeholders as the continue text.",
+  "default.continueTextMaxTokens": LOCALIZED_TEXT_DEFAULTS.en.continueTextMaxTokens,
   "field.guardTools": "Idempotency guard",
   "field.guardToolsHint": "Before resuming, inspect the last tool call: if its result is unconfirmed, tell the model to check state first; if it succeeded, tell it not to rerun — avoids duplicate commits / API calls.",
   "field.guardPendingText": "Guard text (unconfirmed result)",
   "field.guardPendingTextHint": "Appended when the last tool may have partially executed; supports the {tool} placeholder.",
+  "default.guardPendingText": LOCALIZED_TEXT_DEFAULTS.en.guardPendingText,
   "field.guardDoneText": "Guard text (tool succeeded)",
   "field.guardDoneTextHint": "Appended when the last tool is confirmed done; supports {tool} and {result} (result excerpt).",
+  "default.guardDoneText": LOCALIZED_TEXT_DEFAULTS.en.guardDoneText,
   "field.graceMs": "Grace period (ms)",
   "field.graceMsHint": "Wait after an interruption; cancelled if the host recovers on its own.",
   "field.cooldownMs": "Cooldown (ms)",
@@ -179,6 +232,7 @@ var en = {
   "field.loopToolRepeatHint": "How many consecutive calls of the same tool with identical arguments and results trip the loop guard; a changed argument or result counts as progress.",
   "field.loopText": "Loop text",
   "field.loopTextHint": "Text sent after the loop guard restarts a turn; supports the {tool} placeholder.",
+  "default.loopText": LOCALIZED_TEXT_DEFAULTS.en.loopText,
   "stats.byCode": "By error code",
   "stats.empty": "No auto-continue activity today.",
   "stats.reset": "Reset",
@@ -206,37 +260,6 @@ var en = {
 // src/client/settings-card.tsx
 var import_react = require("react");
 
-// src/shared/core.ts
-var DEFAULT_CONFIG = {
-  continueText: "继续",
-  continueTextMaxTokens: "继续",
-  guardTools: true,
-  guardPendingText: "(上一步工具「{tool}」可能未完成, 先确认状态再继续, 不要重复执行)",
-  guardDoneText: "(上一步工具「{tool}」已完成, 结果: {result}; 不要重复执行, 直接继续)",
-  graceMs: 3e3,
-  cooldownMs: 2e4,
-  maxConsecutive: 3,
-  scanOnBoot: true,
-  scanLimit: 8,
-  freshMs: 15 * 60 * 1e3,
-  verbose: true,
-  classify: true,
-  retryableErrorPatterns: "",
-  backoffFactor: 2,
-  backoffMaxMs: 3e5,
-  notify: false,
-  paused: false,
-  loopGuard: true,
-  loopShortChars: 40,
-  loopWindowMs: 3e4,
-  loopShortCount: 12,
-  loopRepeatText: 4,
-  loopToolRepeat: 5,
-  loopText: "(检测到你可能陷入循环, 请停止重复刚才的动作, 换一种方式继续)"
-};
-var RECOVERY_WINDOW_MS = 10 * 60 * 1e3;
-var ECHO_WINDOW_MS = 10 * 60 * 1e3;
-
 // src/client/dsh-store-compat.ts
 function resolveSnapshotStore() {
   const current = ["@deepseek-ai/dsh-client", "-store"].join("");
@@ -1065,7 +1088,7 @@ function AutoContinueSettingsCard(props) {
             ...shared,
             ...state2.continueText,
             onEdit: (text) => props.edit("continueText", text),
-            placeholder: DEFAULT_CONFIG.continueText,
+            placeholder: t("default.continueText"),
             onReset: () => props.resetField("continueText")
           }
         ),
@@ -1078,7 +1101,7 @@ function AutoContinueSettingsCard(props) {
             ...shared,
             ...state2.continueTextMaxTokens,
             onEdit: (text) => props.edit("continueTextMaxTokens", text),
-            placeholder: DEFAULT_CONFIG.continueTextMaxTokens,
+            placeholder: t("default.continueTextMaxTokens"),
             onReset: () => props.resetField("continueTextMaxTokens")
           }
         ),
@@ -1103,7 +1126,7 @@ function AutoContinueSettingsCard(props) {
             ...shared,
             ...state2.guardPendingText,
             onEdit: (text) => props.edit("guardPendingText", text),
-            placeholder: DEFAULT_CONFIG.guardPendingText,
+            placeholder: t("default.guardPendingText"),
             onReset: () => props.resetField("guardPendingText")
           }
         ),
@@ -1116,7 +1139,7 @@ function AutoContinueSettingsCard(props) {
             ...shared,
             ...state2.guardDoneText,
             onEdit: (text) => props.edit("guardDoneText", text),
-            placeholder: DEFAULT_CONFIG.guardDoneText,
+            placeholder: t("default.guardDoneText"),
             onReset: () => props.resetField("guardDoneText")
           }
         ),
@@ -1359,7 +1382,7 @@ function AutoContinueSettingsCard(props) {
             ...shared,
             ...state2.loopText,
             onEdit: (text) => props.edit("loopText", text),
-            placeholder: DEFAULT_CONFIG.loopText,
+            placeholder: t("default.loopText"),
             onReset: () => props.resetField("loopText")
           }
         ),
@@ -1375,8 +1398,18 @@ var SETTINGS_NS = "auto-continue";
 var inject = ["slots", "locale", "settingsScope"];
 function apply(ctx) {
   ctx.effect(() => ctx.locale.register(NS, { zh, en }), "auto-continue: dictionaries");
-  ctx.effect(() => startBridge(), "auto-continue: host bridge");
   const scope = ctx.settingsScope.bind({ namespace: SETTINGS_NS });
+  const syncLocale = () => {
+    const active = ctx.locale.getLocale().active;
+    const snapshot = scope.getSnapshot();
+    if (snapshot.status !== "ready" || !snapshot.writable || snapshot.mode !== "host") return;
+    if (snapshot.value?.locale === active) return;
+    void scope.set("locale", active);
+  };
+  ctx.effect(() => scope.subscribe(syncLocale), "auto-continue: locale settings sync");
+  ctx.on("locale/change", syncLocale);
+  syncLocale();
+  ctx.effect(() => startBridge(), "auto-continue: host bridge");
   const controller = new AutoContinueSettingsCardController(scope);
   ctx.slots.inject(
     "settings.plugin.item",