dsh-client-auto-continue 0.11.0 → 0.11.2

package/README.md

@@ -70,11 +70,17 @@ On host boot it also scans the live sessions: a session whose last turn ended wi
 
 The browser half is a thin shell: the settings card, plus a status bridge that shows notifications (with Resume now / Pause this session 1h buttons, routed back to the host engine) and feeds the card's stats / paused-sessions panels.
 
+### Recovery workflow
+
+The diagram summarizes the automatic recovery path, the loop-guard restart path, and the exit to human intervention. Click it to open the full-size version.
+
+[![dsh-auto-continue recovery workflow](docs/auto-continue-workflow.en.svg)](docs/auto-continue-workflow.en.svg)
+
 ## Quick Start
 
 DSH plugins install into a **profile** (`dsh web` → `web` profile). Install, restart `dsh web`, done.
 
-> **Use the latest DSH (recommended: 0.1.2-alpha.1 or newer).** Run `dsh --version` before installing. Plugin v0.8.2 supports the new 0.1.2 client-store module layout and retains a fallback for DSH 0.1.0-rc.7 through 0.1.1; rc.6 and earlier remain unsupported (`list slot ... requires options.id`). Preview releases may appear on the [official DSH releases page](https://github.com/deepseek-ai/deepseek-harness/releases) before the public npm tag catches up.
+> **Use the latest DSH (recommended: 0.1.2-alpha.4 or newer).** Run `dsh --version` before installing. Plugin v0.11.1 supports the settings API used by DSH 0.1.2-alpha.2+ (including alpha.3 and alpha.4) while retaining compatibility with DSH 0.1.0-rc.7 through 0.1.1; rc.6 and earlier remain unsupported (`list slot ... requires options.id`). Preview releases may appear on the [official DSH releases page](https://github.com/deepseek-ai/deepseek-harness/releases) before the public npm tag catches up.
 
 ### From npm (recommended)
 
@@ -146,9 +152,9 @@ dsh web
 
 ## Configuration
 
-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).
+Everything is configurable from the GUI — no file or console edits needed. Open **Settings → Plugins → Plugin configuration**. **Auto Continue** appears as a collapsed card alongside the other plugins; click the card or its right-hand chevron to expand the full configuration in place. Besides the fields below, the expanded 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).
 
-The **Continuity Relay** card groups controls by handoff, safety, recovery, loop breaking, and live status. Its header also keeps the open-source repository and a **Star on GitHub** shortcut within reach.
+The settings card groups controls by handoff, safety, recovery, loop breaking, and live status. Its header also keeps the open-source repository and a **Star on GitHub** shortcut within reach.
 
 **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.
 
@@ -186,7 +192,7 @@ auto-continue:
 
 **How the card works:**
 
-![Stats & paused sessions](docs/screenshots/07-card-panels.png)
+![The collapsed Auto Continue card in the plugin configuration list](docs/screenshots/02-settings-card.png)
 
 - Edits are **staged** — nothing reaches the disk until you hit **Save**; an unsaved badge marks the card while drafts are pending, and **Discard** drops them
 - A field you changed shows an **Overridden** badge with a per-field **Reset to default** button that restores the built-in value

package/README.zh.md

@@ -70,11 +70,17 @@
 
 浏览器半侧是瘦壳: 设置卡片 + 一条状态桥(展示通知, 带「立即续跑 / 暂停该会话 1 小时」按钮并把动作回传给宿主引擎; 驱动卡片里的统计与暂停面板)。
 
+### 恢复流程
+
+下图汇总了自动恢复主链、循环守卫重启路径和等待人工介入的出口。点击图片可查看原尺寸版本。
+
+[![dsh-auto-continue 恢复流程](docs/auto-continue-workflow.svg)](docs/auto-continue-workflow.svg)
+
 ## 快速开始
 
 DSH 插件安装进 **profile**(`dsh web` 对应 `web` profile)。安装后重启 `dsh web` 即可。
 
-> **请使用最新 DSH(推荐 0.1.2-alpha.1 或更高版本)。** 安装前先运行 `dsh --version`。插件 v0.8.2 已支持 0.1.2 的新 client-store 模块布局, 同时为 DSH 0.1.0-rc.7 至 0.1.1 保留兼容回退; rc.6 及更早版本仍不支持(`list slot ... requires options.id`)。预览版可能先发布在 [DSH 官方 Releases](https://github.com/deepseek-ai/deepseek-harness/releases), 公开 npm 标签会稍后跟进。
+> **请使用最新 DSH(推荐 0.1.2-alpha.4 或更高版本)。** 安装前先运行 `dsh --version`。插件 v0.11.1 已支持 DSH 0.1.2-alpha.2+ 使用的设置 API(包括 alpha.3 和 alpha.4), 同时继续兼容 DSH 0.1.0-rc.7 至 0.1.1; rc.6 及更早版本仍不支持(`list slot ... requires options.id`)。预览版可能先发布在 [DSH 官方 Releases](https://github.com/deepseek-ai/deepseek-harness/releases), 公开 npm 标签会稍后跟进。
 
 ### 从 npm 安装(推荐)
 
@@ -146,9 +152,9 @@ dsh web
 
 ## 配置
 
-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)。
+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 → 插件配置**, **自动继续**会像其他插件一样显示为列表中的折叠卡片; 点击卡片或右侧箭头即可在当前位置展开完整配置。除了下面的字段, 展开后的卡片还带一个**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)。
 
-新版**中断接力器**会按接力方式、安全节奏、恢复雷达、循环断路器与现场状态组织配置; 卡片顶部也直接放出了开源仓库和 **Star on GitHub** 入口。
+新版配置卡会按接力方式、安全节奏、恢复雷达、循环断路器与现场状态组织配置; 卡片顶部也直接放出了开源仓库和 **Star on GitHub** 入口。
 
 **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。
 
@@ -186,7 +192,7 @@ auto-continue:
 
 **卡片操作说明:**
 
-![统计与暂停会话](docs/screenshots/07-card-panels.png)
+![插件配置列表中的自动继续折叠卡片](docs/screenshots/02-settings-card.png)
 
 - 修改是**暂存式**的——点「保存」之前不会写入磁盘; 有待保存草稿时卡片显示「未保存」徽章, 「放弃」可丢弃草稿
 - 改动过的字段会带「已覆盖」徽章, 并有逐字段的「恢复默认」按钮(回到内置默认值)

package/docs/auto-continue-workflow-icons.LICENSE.txt

@@ -0,0 +1,40 @@
+Lucide Icons
+https://lucide.dev/
+
+ISC License
+
+Copyright (c) 2026 Lucide Icons and Contributors
+
+Permission to use, copy, modify, and/or distribute this software for any
+purpose with or without fee is hereby granted, provided that the above
+copyright notice and this permission notice appear in all copies.
+
+THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+
+The Radio icon is derived from Feather Icons and is used under the MIT License:
+
+Copyright (c) 2013-present Cole Bemis
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/docs/auto-continue-workflow.archify.json

@@ -0,0 +1,155 @@
+{
+  "schema_version": 2,
+  "diagram_type": "workflow",
+  "meta": {
+    "title": "dsh-auto-continue 插件流程",
+    "locale": "zh-CN",
+    "animation": "trace",
+    "visual_preset": "signal-flow",
+    "quality_profile": "showcase",
+    "views": [
+      {
+        "id": "auto_recovery",
+        "label": "自动恢复主链",
+        "focus": ["turn_end", "classify_failure", "safety_gate", "grace_wait", "compose_prompt", "agent_followup", "recovery_feedback"],
+        "note": "从非人为中断判定，到安全调度、自动续跑与恢复记账。"
+      },
+      {
+        "id": "loop_recovery",
+        "label": "循环守卫",
+        "focus": ["turn_end", "agent_cancel", "safety_gate", "compose_prompt", "agent_followup"],
+        "note": "检测空转后取消当前回合，并用 loopText 重新进入受控恢复链。"
+      },
+      {
+        "id": "human_feedback",
+        "label": "人工介入与反馈",
+        "focus": ["safety_gate", "manual_stop", "recovery_feedback"],
+        "note": "查看不自动继续的出口，以及浏览器端的状态、通知与操作入口。"
+      }
+    ],
+    "legend": {
+      "mode": "auto",
+      "entries": {
+        "messagebus": { "label": "会话事件" },
+        "backend": { "label": "宿主处理" },
+        "security": { "label": "安全判定" },
+        "cloud": { "label": "Agent 队列" },
+        "frontend": { "label": "浏览器反馈" },
+        "external": { "label": "人工介入" }
+      }
+    },
+    "output": "docs/auto-continue-workflow.html"
+  },
+  "lanes": [
+    { "id": "session", "label": "DSH 会话与浏览器" },
+    { "id": "engine", "label": "AutoContinueRunner" },
+    { "id": "agent", "label": "Agent 执行与中断", "variant": "exception" }
+  ],
+  "groups": [
+    { "id": "host_core", "label": "安全恢复链", "lane": "engine", "fromCol": 1, "toCol": 4, "variant": "emphasis" }
+  ],
+  "mainPath": [
+    "turn_end",
+    "classify_failure",
+    "safety_gate",
+    "grace_wait",
+    "compose_prompt",
+    "agent_followup",
+    "recovery_feedback"
+  ],
+  "semanticChecks": {
+    "allowedRoots": ["turn_end"],
+    "allowedTerminals": ["manual_stop", "recovery_feedback"],
+    "requiredPaths": [
+      { "from": "turn_end", "to": "agent_followup" },
+      { "from": "agent_cancel", "to": "agent_followup" },
+      { "from": "agent_followup", "to": "recovery_feedback" }
+    ]
+  },
+  "nodes": [
+    {
+      "id": "turn_end",
+      "lane": "session",
+      "col": 0,
+      "type": "messagebus",
+      "label": "收到会话信号",
+      "sublabel": "turn/end 或启动扫描"
+    },
+    {
+      "id": "classify_failure",
+      "lane": "engine",
+      "col": 1,
+      "type": "backend",
+      "label": "识别中断原因",
+      "sublabel": "错误、令牌上限与扫描"
+    },
+    {
+      "id": "safety_gate",
+      "lane": "engine",
+      "col": 2,
+      "type": "security",
+      "label": "检查恢复条件",
+      "sublabel": "分类、暂停、冷却、上限"
+    },
+    {
+      "id": "grace_wait",
+      "lane": "engine",
+      "col": 3,
+      "type": "backend",
+      "label": "等待宽限期",
+      "sublabel": "新回合或用户消息会取消"
+    },
+    {
+      "id": "compose_prompt",
+      "lane": "engine",
+      "col": 4,
+      "type": "backend",
+      "label": "组装续跑消息",
+      "sublabel": "模板 + 工具幂等护栏"
+    },
+    {
+      "id": "agent_followup",
+      "lane": "agent",
+      "col": 5,
+      "type": "cloud",
+      "label": "agent.followup",
+      "sublabel": "进入 Agent inbox",
+      "tag": "排队发送"
+    },
+    {
+      "id": "recovery_feedback",
+      "lane": "session",
+      "col": 5,
+      "type": "frontend",
+      "label": "恢复与反馈",
+      "sublabel": "清零、退避、SSE 状态"
+    },
+    {
+      "id": "agent_cancel",
+      "lane": "agent",
+      "col": 1,
+      "type": "security",
+      "label": "agent.cancel",
+      "sublabel": "守卫命中，保留 inbox"
+    },
+    {
+      "id": "manual_stop",
+      "lane": "session",
+      "col": 3,
+      "type": "external",
+      "label": "等待人工介入",
+      "sublabel": "停止、永久错误、达上限"
+    }
+  ],
+  "edges": [
+    { "id": "e_event_classify", "from": "turn_end", "to": "classify_failure", "label": "会话事实", "role": "main" },
+    { "id": "e_classify_gate", "from": "classify_failure", "to": "safety_gate", "label": "中断原因", "role": "main", "variant": "emphasis" },
+    { "id": "e_gate_grace", "from": "safety_gate", "to": "grace_wait", "label": "允许恢复", "role": "main", "variant": "emphasis" },
+    { "id": "e_grace_prompt", "from": "grace_wait", "to": "compose_prompt", "label": "宽限期内无新回合", "role": "main", "variant": "emphasis" },
+    { "id": "e_prompt_followup", "from": "compose_prompt", "to": "agent_followup", "label": "排队续跑", "role": "main", "variant": "emphasis" },
+    { "id": "e_followup_result", "from": "agent_followup", "to": "recovery_feedback", "label": "turn/end + SSE", "role": "return" },
+    { "id": "e_loop_cancel", "from": "turn_end", "to": "agent_cancel", "label": "运行中反复输出或工具调用", "role": "branch", "variant": "security" },
+    { "id": "e_cancel_gate", "from": "agent_cancel", "to": "safety_gate", "label": "aborted + loopText", "role": "return", "variant": "security" },
+    { "id": "e_gate_manual", "from": "safety_gate", "to": "manual_stop", "label": "不可自动恢复", "role": "error" }
+  ]
+}

package/docs/auto-continue-workflow.en.archify.json

@@ -0,0 +1,155 @@
+{
+  "schema_version": 2,
+  "diagram_type": "workflow",
+  "meta": {
+    "title": "dsh-auto-continue Plugin Workflow",
+    "locale": "en",
+    "animation": "trace",
+    "visual_preset": "signal-flow",
+    "quality_profile": "showcase",
+    "views": [
+      {
+        "id": "auto_recovery",
+        "label": "Automatic recovery",
+        "focus": ["turn_end", "classify_failure", "safety_gate", "grace_wait", "compose_prompt", "agent_followup", "recovery_feedback"],
+        "note": "Classify a non-human interruption, apply recovery safeguards, resume the agent, and record the outcome."
+      },
+      {
+        "id": "loop_recovery",
+        "label": "Loop guard",
+        "focus": ["turn_end", "agent_cancel", "safety_gate", "compose_prompt", "agent_followup"],
+        "note": "Cancel a spinning turn, preserve loop context, and re-enter the guarded recovery path."
+      },
+      {
+        "id": "human_feedback",
+        "label": "Human action & feedback",
+        "focus": ["safety_gate", "manual_stop", "recovery_feedback"],
+        "note": "Inspect non-recoverable exits and the browser-side status, notification, and action surface."
+      }
+    ],
+    "legend": {
+      "mode": "auto",
+      "entries": {
+        "messagebus": { "label": "Session event" },
+        "backend": { "label": "Host processing" },
+        "security": { "label": "Safety gate" },
+        "cloud": { "label": "Agent queue" },
+        "frontend": { "label": "Browser feedback" },
+        "external": { "label": "Human action" }
+      }
+    },
+    "output": "docs/auto-continue-workflow.en.html"
+  },
+  "lanes": [
+    { "id": "session", "label": "DSH session & browser" },
+    { "id": "engine", "label": "AutoContinueRunner" },
+    { "id": "agent", "label": "Agent execution & interrupts", "variant": "exception" }
+  ],
+  "groups": [
+    { "id": "host_core", "label": "Safe recovery path", "lane": "engine", "fromCol": 1, "toCol": 4, "variant": "emphasis" }
+  ],
+  "mainPath": [
+    "turn_end",
+    "classify_failure",
+    "safety_gate",
+    "grace_wait",
+    "compose_prompt",
+    "agent_followup",
+    "recovery_feedback"
+  ],
+  "semanticChecks": {
+    "allowedRoots": ["turn_end"],
+    "allowedTerminals": ["manual_stop", "recovery_feedback"],
+    "requiredPaths": [
+      { "from": "turn_end", "to": "agent_followup" },
+      { "from": "agent_cancel", "to": "agent_followup" },
+      { "from": "agent_followup", "to": "recovery_feedback" }
+    ]
+  },
+  "nodes": [
+    {
+      "id": "turn_end",
+      "lane": "session",
+      "col": 0,
+      "type": "messagebus",
+      "label": "Session signal",
+      "sublabel": "turn/end or boot scan"
+    },
+    {
+      "id": "classify_failure",
+      "lane": "engine",
+      "col": 1,
+      "type": "backend",
+      "label": "Classify cause",
+      "sublabel": "Error / token / scan"
+    },
+    {
+      "id": "safety_gate",
+      "lane": "engine",
+      "col": 2,
+      "type": "security",
+      "label": "Recovery gate",
+      "sublabel": "Pause / backoff / cap"
+    },
+    {
+      "id": "grace_wait",
+      "lane": "engine",
+      "col": 3,
+      "type": "backend",
+      "label": "Grace wait",
+      "sublabel": "New turn cancels"
+    },
+    {
+      "id": "compose_prompt",
+      "lane": "engine",
+      "col": 4,
+      "type": "backend",
+      "label": "Build prompt",
+      "sublabel": "Template + tool guard"
+    },
+    {
+      "id": "agent_followup",
+      "lane": "agent",
+      "col": 5,
+      "type": "cloud",
+      "label": "agent.followup",
+      "sublabel": "Enter Agent inbox",
+      "tag": "Queued"
+    },
+    {
+      "id": "recovery_feedback",
+      "lane": "session",
+      "col": 5,
+      "type": "frontend",
+      "label": "Restore status",
+      "sublabel": "Reset + SSE state"
+    },
+    {
+      "id": "agent_cancel",
+      "lane": "agent",
+      "col": 1,
+      "type": "security",
+      "label": "agent.cancel",
+      "sublabel": "Guard hit; keep inbox"
+    },
+    {
+      "id": "manual_stop",
+      "lane": "session",
+      "col": 3,
+      "type": "external",
+      "label": "Human action",
+      "sublabel": "Pause / error / cap"
+    }
+  ],
+  "edges": [
+    { "id": "e_event_classify", "from": "turn_end", "to": "classify_failure", "label": "Session facts", "role": "main" },
+    { "id": "e_classify_gate", "from": "classify_failure", "to": "safety_gate", "label": "Failure cause", "role": "main", "variant": "emphasis" },
+    { "id": "e_gate_grace", "from": "safety_gate", "to": "grace_wait", "label": "Recovery allowed", "role": "main", "variant": "emphasis" },
+    { "id": "e_grace_prompt", "from": "grace_wait", "to": "compose_prompt", "label": "No new turn", "role": "main", "variant": "emphasis" },
+    { "id": "e_prompt_followup", "from": "compose_prompt", "to": "agent_followup", "label": "Queue followup", "role": "main", "variant": "emphasis" },
+    { "id": "e_followup_result", "from": "agent_followup", "to": "recovery_feedback", "label": "turn/end + SSE", "role": "return" },
+    { "id": "e_loop_cancel", "from": "turn_end", "to": "agent_cancel", "label": "Repeated output or tool calls", "role": "branch", "variant": "security" },
+    { "id": "e_cancel_gate", "from": "agent_cancel", "to": "safety_gate", "label": "aborted + loopText", "role": "return", "variant": "security" },
+    { "id": "e_gate_manual", "from": "safety_gate", "to": "manual_stop", "label": "Manual action required", "role": "error" }
+  ]
+}