feat: support native subagents and long-running prompts

OpenCode Task calls now appear as first-class subagents in local and managed SSH conversations, with stable status, persistence, terminal handling, and true event ordering instead of duplicate generic tool entries.

GoodBuddy no longer applies a fixed outer wall-clock total to OpenCode, Continue, DeepSeek Harness, or managed SSH Prompts. Runtime-owned limits, user cancellation, bounded control operations, output limits, and released Agent protocol compatibility remain intact. Desktop and Agent release metadata are synchronized at 0.11.13 with approved bilingual notes.

Release note: OpenCode 原生 Task 现在以独立子 Agent 展示；长任务不再被 GoodBuddy 固定总时限中止，并修复 OpenCode 启动识别、可配置 HTTP 服务地址与发布说明读取问题。

AGENTS.md

@@ -265,6 +265,39 @@ possible, and report the exact call count. This standing authorization does
 not cover bulk, high-cost, destructive, publishing, messaging, purchasing, or
 other consequential external actions.
 
+### GoodBuddy Agent development validation
+
+Every source change that can alter the deployed GoodBuddy Agent or the
+desktop-to-Agent production path must be validated on the shared Linux x64 test
+Host while the change is being developed. This includes Agent daemon and
+protocol changes, packaging and installation, attach/update behavior, Runtime
+launch and process ownership, workspaces, model bridges, and lifecycle or
+recovery behavior. Unit tests, mocks, fixtures, CI, and waiting for release
+validation are not substitutes for this development-time check.
+
+- Reach the shared Host at `192.168.0.23` on the LAN when available, otherwise
+  use `10.7.0.23` over VPN. These are two routes to the same physical machine,
+  not two independent test targets or architecture results.
+- Exercise the current source or package under development, not only the
+  previously published Agent. Run the affected production path as soon as it is
+  runnable and repeat the relevant scenario after the final change.
+- Select the real-Host scenario from the changed surface: attach/bootstrap or
+  update for installation changes; Ask/Execute and a bounded real model request
+  for Runtime or model-bridge changes; and the applicable disconnect, restart,
+  cancellation, or reconnection sequence for lifecycle and recovery changes.
+- Use the existing pinned Host identity and GoodBuddy credential storage. Never
+  add credentials to source, documentation, commands, logs, or test output, and
+  never hard-code these test addresses into product behavior or defaults.
+- Keep all artifacts in GoodBuddy-owned test directories. Before stopping or
+  changing a remote process, verify that it belongs to the test run; never
+  modify unrelated Host files or processes.
+- If neither route is reachable, report the real-Host validation as blocked and
+  do not present the Agent work as complete or defer its first real validation
+  to release time.
+
+The detailed scenario rules are documented in
+[`docs/architecture/remote-host-and-goodbuddy-agent-design.md`](./docs/architecture/remote-host-and-goodbuddy-agent-design.md).
+
 Before committing or pushing, inspect `git status`, `git diff`, and
 `git diff --cached`. Do not commit secrets, local databases, logs, generated
 credentials, or private user artifacts.

UI-DESIGN.md

@@ -553,6 +553,7 @@ GoodBuddy 是可调整窗口大小的桌面应用。响应式设计优先保证
 - OpenCode、Continue 和 DeepSeek Harness 后续的 Task 级委派与取消、后台执行进度/结果、Workflow/Hook 运行、长任务暂停/恢复/终止及原生会话监督统一进入应用级助手工作栏固定的“Runtime”栏目，不加入 Composer。栏目入口始终存在；只选择 Conversation 或 Task，并按 Task 聚合执行状态，不显示 Job/Run 树或独立操作菜单。内部可选区域按用户所选目标和 Runtime 的真实能力显示，不为未支持能力渲染空卡片或成排禁用按钮。切换跟随目标时必须清理上一归属的监督状态，固定目标则保持不变。
 - 设置中心只管理持久 Runtime 配置、默认值和能力清单；右侧 Runtime 栏目管理用户当前跟随或固定目标的生命周期。两处不得复制同一实时操作，栏目中的高风险操作仍须就地确认并保留取消、权限、用量和活动审计。
 - Runtime Prompt 快捷操作只把模板填入输入草稿，用户可以继续编辑；OpenCode Command 由 Runtime 原生 API 执行，输入框只承载可选参数，不以普通斜杠文本冒充执行。
+- OpenCode 原生 Task 调用在本地项目和托管 SSH 项目中统一解析为子 Agent 状态卡，不显示为普通工具卡。状态卡持续显示 Agent 类型、任务说明、运行状态、完成输出或失败原因；身份徽标、Agent 名称和状态位于首层，任务说明作为主信息，Runtime 来源作为次级元数据，输出与错误在同一卡片内按需展开。父 Agent 的推理不冒充子 Agent 过程，必须与状态卡保持明确间距。状态不能只靠颜色表达。推理、子 Agent、工具和正文统一按事件首次出现顺序进入消息块；增量更新只更新原位置，Task 临时工具块重分类后必须原位替换，不能把子 Agent 卡统一提升到消息顶部。同一调用的增量事件按稳定调用 ID 更新，不能因 ACP 首帧尚无参数而永久保留重复工具卡或活动记录。一次请求内已观察到的子 Agent 均保留在消息与活动记录中；父请求取消、失败或意外中断时仍处于活动状态的卡片必须结算为对应终态，父请求成功结束但子 Agent 未报告终态时必须显示为失败而不是推断为完成。
 - Agent 回复进行中锁定模型、专家角色、工作模式和 Runtime 定制选择器，并关闭已打开的上下文菜单；回复结束或停止后再恢复选择，避免界面状态与本次运行实际使用的上下文不一致。
 - Agent 回复进行中不能禁用普通消息发送。新消息与到期 Scheduled Task 共用 Conversation 级待发送队列。队列位于 Composer 容器之外、与输入框等宽并保持独立间距；有待发送项时直接显示无标题栏的极简列表，空队列不保留占位。每项固定为一行，按进入顺序显示轻量来源图标和截断摘要；普通消息使用消息图标，Scheduled Task 统一使用淡出时钟图标。
 - 待发送项默认在当前回复结束后顺序执行。每项提供明确的“立即中断并插入”（空闲时为“立即运行”）和删除操作；前者先取消当前 Conversation 的活动执行，再将所选项提升为下一项，不允许同一 Conversation 并发写入时间线。
@@ -626,6 +627,7 @@ GoodBuddy 是可调整窗口大小的桌面应用。响应式设计优先保证
 - “笔记 / 待办”属于同一工作台内的同级内容面板，使用 `PageTabs` 的 `segmented` 视觉变体，与模型设置的分段控件保持同一外观。
 - 页签切换保留 `tablist`、`tab` 和 `tabpanel` 语义；待办状态仍使用独立的 `SegmentedControl`，不得与内容页签合并。
 - 笔记编辑器保留 Quill 原生列表快捷输入：输入 `-`、`*`、`1.`、`[ ]` 或 `[x]` 后按空格，分别转换为项目符号、编号或待办清单。快捷输入只生成现有 Delta 富文本，不引入 Markdown 源码模式、第二套持久化格式或第三方 Markdown 插件；编辑区附近持续提供简短提示。
+- 左侧笔记列表依次显示标题、最多两行内容预览，以及同一行的记录数和本地化更新时间。记录数靠左、更新时间靠右，窄宽度下均保持单行省略，不挤压标题或扩大列表栏。
 - 左侧“魔法笔记”入口可显示未完成待办数量。数量为 0 时隐藏徽标，超过 99 时显示 `99+`，可访问名称仍说明精确数量；勾选、取消勾选以及 Agent / MCP 对笔记清单的写入都必须刷新数量。
 - 未完成待办徽标的高度不得超过主导航图标；数量从 0 变为可见或再次隐藏时，主导航项行高和相邻菜单位置必须保持不变。
 - 在待办面板勾选或取消勾选后，当前缓存中的源笔记正文必须立即同步对应清单状态，不得依赖离开工作区后重新加载。

agent-runtime-lock.json

@@ -1,6 +1,6 @@
 {
   "formatVersion": 1,
-  "agentVersion": "0.11.9",
+  "agentVersion": "0.11.13",
   "protocol": {
     "major": 2,
     "minor": 0

docs/architecture/deepseek-harness-runtime-design.md

@@ -163,7 +163,7 @@ Electron utilityProcess
        ├─ 本地 Shell / Filesystem Provider
        ├─ 最小工具集与 Main 代理 MCP
        └─ Main 明确启用的第三方 Cordis 插件
-             │ HTTPS
+             │ HTTP/HTTPS
              ▼
         用户选择的 OpenAI 兼容模型连接
 ```
@@ -325,12 +325,13 @@ GoodBuddy conversationId -> Harness sessionId + process generation
 - 历史以明确的“不可信会话数据”结构传入，不能拼接成系统指令。
 - 用户分配的 Skill 只通过 Main 校验的包路径进入 Host，并在 Agent scope 注册；不得把 Skill 内容伪装成用户 Prompt。
 
-### 9.3 取消与超时
+### 9.3 取消与有界控制操作
 
 - 用户取消时立即发送 `session/cancel`。
 - 取消等待有界，超时后关闭连接并终止整个 Harness 进程。
-- 初始化、握手、Session 创建、Prompt、权限回传和关闭分别使用独立超时。
+- 初始化、握手、请求准备、权限回传和关闭分别使用独立控制超时。
-- Prompt 超时与用户取消使用不同错误类型，不能被宽泛 catch 抹平。
+- 生产 Prompt 不设置固定墙钟总时限，由 Harness 自身协议终态或用户取消结束。测试可以
+  显式注入短 Prompt 时限验证取消传播，超时错误与用户取消仍不能被宽泛 catch 抹平。
 - 取消后仍可接收并丢弃该请求的最终协议结算帧，但不得写入下一请求。
 
 ### 9.4 释放与退出
@@ -426,8 +427,7 @@ DeepSeek Harness 首版只使用符合下列边界的 GoodBuddy 模型连接：
 
 - 协议必须是 `openai-chat-completions`。
 - 认证必须是 API Key。
-- 公网服务地址必须使用 HTTPS；`localhost`、`127.0.0.1` 和 `::1` 回环地址可以使用 HTTP。
+- 服务地址支持 HTTP 和 HTTPS，不根据主机位置、端口、凭据、查询参数或片段增加额外限制。
-- 服务地址可以使用自定义主机、端口和部署路径，但不得包含用户名、密码、查询参数或片段。
 - 模型名称不限制为 DeepSeek 品牌，由所选 OpenAI 兼容服务决定。
 - 模型名称和服务地址由 Main 传入受控 Host。
 - 图片能力只读取所选 GoodBuddy 模型连接的 `supportsImageInput`；Main、Utility 启动配置、ACP 能力和 Pi-AI 模型输入模态必须使用同一个布尔值。
@@ -469,7 +469,7 @@ DeepSeek Harness 首版只使用符合下列边界的 GoodBuddy 模型连接：
 | stderr 累计 | 64 KiB |
 | Host 启动 | 10 秒基础预算 + 每插件 5 秒激活与最多 1 秒失败清理，插件序列最多 91 秒；Main 另预留 2 秒持久化失败状态 |
 | ACP 初始化与内部握手 | 每阶段 10 秒 |
-| 单次 Prompt | 10 分钟 |
+| 单次 Prompt | 无固定墙钟总时限，由 Harness 终态或用户取消结束 |
 | 有序关闭宽限期 | 2 秒 |
 
 超过限制时应取消当前请求。协议帧、队列或 stderr 持续异常时，应终止 Runtime 进程，避免继续信任已失控的通道。

docs/architecture/remote-host-and-goodbuddy-agent-design.md

@@ -2,7 +2,7 @@
 
 ## 状态
 
-本文记录截至 2026-08-29 的当前代码实现，不定义额外的信任框架。“新增 Host 只探测、Host 卡片手动准备
+本文记录截至 2026-08-30 的当前代码实现，不定义额外的信任框架。“新增 Host 只探测、Host 卡片手动准备
 Agent/Runtime、Host 直接从 GitHub/北京镜像下载、项目始终使用 Host current 环境”已经完成源码接线，
 详细事务与验收边界见
 [SSH Host 远程环境准备与直连下载设计](./remote-host-environment-provisioning-design.md)；
@@ -13,8 +13,9 @@ Windows 到 Linux x64 的既有安装、
 Main-only 模型桥、断线恢复、输出重放、同一 OpenCode Session 续接与终态清理已经在真实
 Host 上完成 provider-free 验证。上一轮 Linux x64 验收使用 Agent `0.11.2-e2e.12` 与
 OpenCode Runtime `1.18.9` 完成了一次有界真实模型工具调用；加入每 helper 随机 loopback
-路径 capability 后，Agent `0.11.9` 已通过独立 workflow 发布 Linux x64/arm64 复合包和
+路径 capability 后，Agent `0.11.10` 已通过独立 workflow 发布 Linux x64/arm64 复合包和
-签名累计目录；当前桌面源码准备 `0.11.10`，修复 Host 更新并消除重复 payload 验证。失败的
+签名累计目录；当前桌面源码准备 `0.11.13`，显示本地与远端 OpenCode 原生 Task，并取消
+GoodBuddy 对生产 Prompt 的固定墙钟总时限。失败的
 `agent-v0.11.3` 保持不可变且未发布。
 
 ## 产品语义
@@ -193,7 +194,13 @@ Execute 不经过 Ask 的 bubblewrap profile：
 - 继承 SSH 账号的正常环境、文件系统、进程和网络能力；
 - 不进行 T2/T3、confinement attestation、approval bridge 或逐工具批准。
 
-两种模式都保留请求 deadline、输入/输出字节上限、取消和进程组清理。
+两种模式都保留输入/输出字节上限、用户取消和进程组清理。生产 Prompt 不设置固定墙钟
+总时限；只要 Runtime 尚未按自身协议结束，Main 和 Agent 就允许其持续运行。启动、握手、
+单个控制 RPC、重连尝试和关闭清理仍使用独立的有界超时，测试可以显式注入短 Prompt
+时限验证取消升级，但该覆盖值不是生产默认。无界 Prompt 继续复用已发布的必填
+`deadlineAt` 字段，以 year-9999 哨兵表达；Main 与 Agent 必须同时理解该哨兵，Agent
+`0.11.10` 及更早版本仍会按其 manifest 上限夹紧，因此桌面启用这一语义前必须配套更新
+到 Agent `0.11.13` 或更高版本。
 
 ## ACP 与断线
 
@@ -207,13 +214,21 @@ Execute 不经过 Ask 的 bubblewrap profile：
   调度、跨 channel 超车、批发送或第二套拥塞控制。
 - Agent 在把 Main 输入交给 Runtime 前、以及把 Runtime 输出交给 Main 前，先把 ACP frame 写入本机 journal。ACK 只推进 cursor 并裁剪已确认 frame，不表示 channel 已终止。
 - Main 持久化 binding identity 和单调 cursor；保留中的 connection lease 即使处于 offline/reconnecting，也可以先把新 cursor 落盘。重连提交不能覆盖 resume 过程中并发落盘的更新。
-- Runtime 事件进入桌面会话后，Renderer 和 Main 将完整工具详情保存到本地 SQLite，不再对已接受的工具输入、输出、错误或摘要施加额外的会话字段长度限制。
+- Runtime 事件先经过单事件、单请求输出和每次 Prompt 最多 100 个不同工具调用的边界；
+  通过边界后，Renderer 和 Main 将已接受的工具详情保存到本地 SQLite，不再额外缩短其
+  输入、输出、错误或摘要。
+- OpenCode 原生 Task 工具按 `subagent_type`、`description`、`prompt` 和稳定 tool call ID
+  解析为子 Agent 事件。本地 OpenCode SDK 与远端 ACP 增量工具事件复用同一转换；ACP
+  首帧缺少参数时可以先显示普通工具活动，后续参数确认其为 Task 后必须替换为子 Agent
+  状态卡，并持久化任务说明、终态与输出。
 - 同一 detached Agent 存活时，短暂 SSH 断线依次执行 `controller/resume`、`runtime/resumeAcpChannel` 和 `runtime/replayAcpChannel`，从 Main 已确认的 cursor 后只重放 Agent 到 Main 的已记录输出。重复 frame 会被确认但不会再次交给上层。若上一代连接只留下没有活动请求的 detached binding，完成精确 controller takeover 后会先有界停止并核对遗留 Runtime process，再用新 channel epoch 重新打开同一 binding 并恢复已有 ACP session；其他 controller、未证明 takeover 或仍有活动请求的 binding 仍被拒绝。
 - Main 到 Runtime 的 ACP 输入、模型请求、工具请求和 blob 不自动重放。
 - 无法确认外部 Provider 是否已处理的模型调用保持结果未知，避免重复计费或重复副作用。
 - 只有远端返回 identity 匹配且 `closed: true` 时，Main 才删除持久化 binding；传输失败或未确认 close 会保留 recovery identity。
 - 终态 close 由 Agent 在一个事务中为两个方向记录 sequence high-water tombstone，删除剩余 frame 和 active channel，并归零对应 journal quota。tombstone 用于拒绝迟到的旧 epoch frame，因此当前不会按时间自动裁剪。
-- 断线不主动终止正在运行的 Runtime。用户取消、deadline、显式关闭或 Agent shutdown 才触发停止。
+- 断线不主动终止正在运行的 Runtime。用户取消、显式关闭、输出边界或 Agent shutdown
+  才触发停止；连接恢复只受当前用户取消信号和单次重连控制超时约束，不把 Prompt 已运行
+  时长当作停止条件。
 - Agent 失联期间继续把 Runtime 输出写入本机 journal；重新连接只校验 controller
   identity、恢复 binding，并从 Main 最后确认的 cursor 后同步缺失数据。已接受的
   指令和结果不确定的 Provider 请求都不自动重放。
@@ -224,8 +239,8 @@ Execute 不经过 Ask 的 bubblewrap profile：
 - 远端 OpenCode 通过 GoodBuddy Agent helper 和每次 Prompt 的私有 Unix socket 使用模型桥。
 - Main 校验固定模型 profile、协议、请求路径和有界传输格式，并记录 Provider
   返回的实际 usage；不限制 Prompt 内的模型调用轮数、工具调用次数、累计 Token
-  或单次模型输出 Token。取消、Runtime deadline、请求/响应字节上限和结果不确定时
+  或单次模型输出 Token。取消、请求/响应字节上限和结果不确定时禁止自动重放仍然保留；
-  禁止自动重放仍然保留。
+  模型桥不会为整个 Runtime Prompt 另设墙钟总时限。
 - helper 可以接收同一 Prompt 内并发到达的模型桥请求；它在单一稳定模型桥上按到达
   顺序等待并交付，不返回本地 `bridge-busy`，每个响应只有在 HTTP 完整 flush 后才
   发送 delivery ACK。
@@ -242,12 +257,55 @@ Execute 不经过 Ask 的 bubblewrap profile：
   不修改既有 React state 对象。这样开发环境 StrictMode 重复调用 state updater 时，
   block metadata 与 canonical 消息正文保持一致。
 
+## Agent 开发期间的真实 Host 验证
+
+任何会改变已部署 GoodBuddy Agent 或桌面到 Agent 生产链路的源码改动，都必须在开发期间
+使用共享 Linux x64 真实测试 Host 验证。适用范围包括 Agent daemon、Agent protocol
+及共享 contract、Agent 组包与安装、attach/update、Runtime 启动与进程归属、Workspace、
+model bridge，以及生命周期和恢复逻辑。单元测试、mock、fixture、CI 或正式发布前回归
+都不能替代这一步；不得等到准备发布时才首次运行受影响的真实链路。
+
+测试 Host 有两个网络入口：
+
+- `192.168.0.23`：局域网入口，在本地网络可达时优先使用；
+- `10.7.0.23`：VPN 入口，局域网入口不可达时使用。
+
+两者指向同一台物理 Host，只代表两条访问路由，不能计为两个独立 Host、两种架构或两份
+覆盖结果。测试只使用已固定的 Host identity 和 GoodBuddy 已有凭据存储；地址不得写入
+产品行为或默认配置，用户名、密码、私钥等凭据不得写入源码、文档、命令、日志或测试输出。
+
+开发时应在改动链路达到可运行状态后尽早执行真实 Host 验证，并在最终相关改动后重复受
+影响的场景。必须使用当前源码构建或当前候选包，不能只验证 Host 上已经发布的旧 Agent。
+按改动范围至少选择以下场景：
+
+1. Agent 组包、安装或升级：验证当前候选的 attach-or-bootstrap、安装/更新、版本切换及
+   随后的连接。
+2. Runtime profile、进程启动/归属或 Ask/Execute 边界：通过桌面生产入口启动真实 Runtime，
+   验证受影响模式；Ask 权限改动必须验证只读边界，Execute 权限改动必须验证所需写入、
+   进程和网络能力。
+3. Agent protocol、RPC、stream、model bridge 或模型消息链路：执行最小有界真实模型调用，
+   禁用非必要工具和附件，并记录实际调用次数；大消息或工具链路改动还应覆盖对应的真实
+   payload/工具场景。
+4. 会话、owner、取消、超时、断线或恢复逻辑：执行与改动直接相关的中断、取消、Agent/
+   Runtime 重启、SSH 断开或重连序列，并确认不会错误重放结果未知的操作。
+5. Workspace 或 Host 文件路径：只在 GoodBuddy 专用测试目录操作，验证真实路径和权限，
+   不得读取、覆盖或删除无关 Host 文件。
+
+停止或修改远端进程前必须先核对其启动身份、PID/进程组和 GoodBuddy 归属。若两个入口均
+不可达，应立即把真实 Host 验证报告为开发阻塞，保持该验证未完成，不能把 Agent 改动表述
+为已完成，也不能把首次真实验证推迟到发布阶段。
+
 ## 已完成的 E2E 验收记录
 
 - 2026-08 的本地 fixture 完整验证 Linux x64 Agent `0.11.2-e2e.12`、Node `24.19.0`
   和 Agent protocol `2.0`；当时没有 arm64 fixture，因此该记录不能作为当前独立发布
-  的双架构验收。Agent `0.11.9` 后续已由原生 workflow 发布并公开验证双架构工件；
+  的双架构验收。Agent `0.11.10` 后续已由原生 workflow 发布并公开验证双架构工件；
-  当前源码 lock 固定为 `0.11.9`。
+  当前源码 lock 固定为候选 `0.11.13`。
+- 2026-08-30 在共享 Linux x64 Host 的隔离测试 HOME 中验证当前 `0.11.13` 源码候选：
+  签名 Runtime 清单的测试墙钟上限为 1 秒，模型桥首轮故意延迟 2.515 秒后 Prompt 仍在
+  5.770 秒正常完成；随后 OpenCode 原生 Task 依次产生 `running`、`completed` 子 Agent
+  事件并完成父请求。测试结束后隔离 Agent、Runtime、socket 和 Workspace 已全部清理，
+  Host current Agent 仍为 `0.11.10`。
 - 正常 Host 更新路径把 Linux x64 Host 的 Agent 更新为 `0.11.2-e2e.12`，并确认
   OpenCode Runtime 已安装版本与所需版本均为 `1.18.9`。
 - 一条新的 Ask 用户操作只提交一次。OpenCode 先在 build 模型轮次请求一个原生
@@ -284,7 +342,12 @@ Workspace、Runtime 和 Saving 进度仅用于新建或显式保存项目；进
 产生冲突的项目创建和设置入口。远端 RPC 拒绝只向
 Renderer 暴露固定方法名、数字 RPC code 和有界 service code，不转发 Host 私有错误详情。
 `runtime/preparePrompt` 在远端明确拒绝且尚未接受 Prompt 时会关闭对应 Main binding，
-不会把确定性失败错误保留为 `outcome-unknown`。
+不会把确定性失败错误保留为 `outcome-unknown`。未确认终态的 Prompt 仍不会自动重放；
+用户之后显式发送的新 Prompt 会关闭旧的非就绪 Main binding，并以新的 ACP session
+继续当前 Conversation，避免一次恢复不确定永久阻塞后续输入。
+Ask 模式的 Linux Runtime 进程由 `bwrap` 启动后 `exec` 到已签名 Runtime 或 Agent
+model bridge helper；监督器只接受同一 PID/进程组从固定 `bwrap` 路径到预期最终可执行文件
+的有界转换，再登记为运行中进程。
 
 不持久化 T2/T3、consent、approval bridge、confinement 或组件验证结果。数据库 schema
 v31 保留项目及关联用户数据，但重建执行空间表为 `project_id/kind/root_path/ssh_host_id`
@@ -314,7 +377,10 @@ v31 保留项目及关联用户数据，但重建执行空间表为 `project_id/
   信任条件。上传到 Host 前重新序列化为确定性 LF JSON。签名目录、描述符、manifest 和
   payload 仍按原始签名字节及 SHA-256 验证。
 
-## 发布前验证
+## 发布前完整回归（不能替代开发验证）
+
+以下清单只用于复核开发期间已经在真实 Host 上通过的链路，不得作为 Agent 改动的首次
+真实 Host 验证：
 
 1. `npm test`
 2. `npm run typecheck`

docs/prd/assistant-experience/assistant-workbar-and-execution-spaces-prd.md

@@ -806,8 +806,9 @@ Electron、ChildProcess、PTY、SSH Client、Socket 或文件句柄。
 - 通用 Artifact Event。
 - HTML 工作区和成果的静态即时预览。
 - Runtime Inspector Snapshot 与事件。
-- OpenCode 会话、子会话、Todo、工具、用量和取消。
+- OpenCode 会话、子会话、Todo、工具、用量和取消；本地与托管 SSH 的原生 Task 调用在
-- 直连模型及现有 GoodBuddy Subagent 的统一展示。
+  Conversation 中统一显示为可追踪的 Subagent 状态，而不是普通工具参数卡。
+- 直连模型、OpenCode 原生 Subagent 及现有 GoodBuddy Subagent 的统一展示。
 
 ### 阶段 2：监督、终端与受管进程
 

docs/roadmap/long-term-assistant-roadmap.md

@@ -266,7 +266,7 @@ GoodBuddy 应能够：
 1. 所有新 IPC 继续执行 Zod 校验和可信主窗口校验。
 2. 项目根目录、成果路径和上下文路径必须 canonicalize 并验证目录包含关系。
 3. Child Runtime 环境变量改用最小 allowlist，避免继承无关密钥。
-4. 远程委派仅允许 HTTPS，开发环境只放行 loopback。
+4. 远程委派支持用户配置的 HTTP 和 HTTPS 地址，不按主机位置增加额外限制。
 5. 语音、窗口捕获和浏览器上下文分别授权。
 6. 自动化不得绕过工具审批和项目权限。
 7. 记忆必须保留来源和作用域。

package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "goodbuddy",
-  "version": "0.11.11",
+  "version": "0.11.13",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "goodbuddy",
-      "version": "0.11.11",
+      "version": "0.11.13",
       "license": "0BSD",
       "dependencies": {
         "@agentclientprotocol/sdk": "0.25.1",

package.json

@@ -1,6 +1,6 @@
 {
   "name": "goodbuddy",
-  "version": "0.11.11",
+  "version": "0.11.13",
   "private": true,
   "description": "Secure desktop AI workspace with controlled Agent Runtimes",
   "desktopName": "GoodBuddy",

resources/release-notes.json

@@ -1,6 +1,78 @@
 {
   "formatVersion": 1,
   "releases": [
+    {
+      "version": "0.11.13",
+      "releasedAt": "2026-08-30",
+      "notes": {
+        "zh-CN": {
+          "highlights": [
+            "GoodBuddy 0.11.13 将本地与托管 SSH 项目中的 OpenCode 原生 Task 显示为独立子 Agent，并移除 GoodBuddy 固定的 Prompt 总时限，让 Runtime 决定长任务生命周期。"
+          ],
+          "features": [
+            "**OpenCode 原生子 Agent。** OpenCode Task 调用现在以独立子 Agent 卡片显示，持续呈现任务、运行状态、输出和错误，不再同时重复为普通工具。"
+          ],
+          "fixes": [
+            "**长任务不再被固定时限中止。** 移除本地 OpenCode、Continue、DeepSeek Harness 和托管 SSH Prompt 的 2/10 分钟外层总时限。任务可持续数小时或更久，直到正常完成或用户取消；Runtime 自身超时仍然生效。",
+            "**OpenCode 启动更可靠。** 改用本机健康接口检测启动完成，不再依赖固定日志格式，并为较慢设备保留更充足的冷启动时间，减少已启动却误报超时。",
+            "**HTTP 服务地址配置更灵活。** 模型连接、DeepSeek Harness 与微信绑定可使用用户明确配置的合法 HTTP 或 HTTPS 地址，包括局域网与公网服务；仍保留协议和 API Key 校验。"
+          ],
+          "notices": [
+            "用户仍可随时停止任务；启动、控制请求、断线恢复、取消收敛、清理和输出大小仍有独立边界。",
+            "托管 SSH Host 需要单独发布的 Agent `0.11.13`。提供 Linux x64 和 arm64；桌面包不内置 Agent 或远程 Runtime。"
+          ]
+        },
+        "en-US": {
+          "highlights": [
+            "GoodBuddy 0.11.13 presents native OpenCode Tasks as dedicated subagents in local and managed SSH projects, and removes GoodBuddy's fixed Prompt deadline so the Runtime owns long-task lifetime."
+          ],
+          "features": [
+            "**Native OpenCode subagents.** OpenCode Task calls now appear as dedicated subagent cards with the task, running state, output, and errors, without a duplicate generic-tool entry."
+          ],
+          "fixes": [
+            "**Long tasks are no longer stopped by a fixed limit.** Removed the outer 2/10-minute total deadline from local OpenCode, Continue, DeepSeek Harness, and managed SSH Prompts. Tasks can run for hours or longer until completion or user cancellation; Runtime-owned timeouts still apply.",
+            "**More reliable OpenCode startup.** GoodBuddy now detects readiness through the local health endpoint instead of a fixed log format and allows more cold-start time on slower devices, avoiding false startup timeouts.",
+            "**More flexible HTTP service endpoints.** Model connections, DeepSeek Harness, and Weixin binding can use explicitly configured valid HTTP or HTTPS endpoints, including LAN and public services, while retaining protocol and API-key validation."
+          ],
+          "notices": [
+            "Users can still stop tasks at any time. Startup, control requests, reconnection recovery, cancellation settlement, cleanup, and output sizes retain independent bounds.",
+            "Managed SSH Hosts require the separately published Agent `0.11.13`. Linux x64 and arm64 are provided; desktop packages do not embed the Agent or Remote Runtime."
+          ]
+        }
+      }
+    },
+    {
+      "version": "0.11.12",
+      "releasedAt": "2026-08-30",
+      "notes": {
+        "zh-CN": {
+          "highlights": [],
+          "features": [
+            "**魔法笔记更新时间。** 魔法笔记列表现在显示每篇笔记的最后更新时间，便于快速判断最近修改过的内容。"
+          ],
+          "fixes": [
+            "**远程会话中断后可以继续输入。** 修复托管 SSH 请求结果不确定后，同一会话的每条后续输入都会重复显示“恢复状态不确定”的问题。未确认的旧请求仍不会自动重放；用户再次明确发送时，GoodBuddy 会建立新的 ACP session 并继续当前会话。",
+            "**Ask 模式远程 Runtime 可以稳定启动。** 修复 Linux Host 上 `bwrap` 启动身份检查存在竞态，导致 Ask 模式可能在模型调用前被拒绝的问题。Agent 现在会验证从固定沙箱启动器到预期签名可执行文件的受限转换。"
+          ],
+          "notices": [
+            "**远程项目技术预览与 Agent 要求。** 远程项目仍默认关闭且为技术预览。Host 需要更新到单独发布的 Agent `0.11.10` 才能获得 Ask 模式启动修复；提供 Linux x64 和 arm64，桌面包不内置 Agent 或远程 Runtime。本次不构建 LoongArch 预览。"
+          ]
+        },
+        "en-US": {
+          "highlights": [],
+          "features": [
+            "**Magic Notes update times.** The Magic Notes list now shows when each note was last updated, making recently edited content easier to identify."
+          ],
+          "fixes": [
+            "**Remote conversations continue after interruptions.** Fixed every later message in a managed SSH conversation repeatedly failing with an uncertain-recovery error after a prompt outcome could not be confirmed. The uncertain prompt is still never replayed automatically; when the user explicitly sends again, GoodBuddy starts a new ACP session and continues the current conversation.",
+            "**Reliable Ask-mode Remote Runtime startup.** Fixed a race in Linux Host `bwrap` launch identity verification that could reject Ask-mode requests before the model was called. The Agent now verifies the bounded transition from the fixed sandbox launcher to the expected signed executable."
+          ],
+          "notices": [
+            "**Remote Projects technical preview and Agent requirement.** Remote Projects remain disabled by default and in technical preview. Hosts must update to the separately published Agent `0.11.10` to receive the Ask-mode startup fix. Linux x64 and arm64 are provided; desktop packages do not embed the Agent or Remote Runtime. This release does not include a LoongArch preview."
+          ]
+        }
+      }
+    },
     {
       "version": "0.11.11",
       "releasedAt": "2026-08-30",

src/agent-daemon/direct-linux-stdio-process-owner.test.ts

@@ -5,6 +5,9 @@ import { join } from 'node:path'
 import { PassThrough } from 'node:stream'
 import { afterEach, describe, expect, it, vi } from 'vitest'
 import type { RemoteRuntimeBundleManifest } from '../shared/remote-runtime-launch-contracts'
+import {
+  UNBOUNDED_REMOTE_PROMPT_DEADLINE
+} from '../shared/remote-agent-contracts'
 import {
   launchDirectLinuxStdioProcessOwner,
   reconcileOrphanedDirectLinuxStdioProcesses,
@@ -119,6 +122,47 @@ describe('direct Linux stdio Runtime ownership', () => {
     registry.close()
   })
 
+  it('waits for bubblewrap to exec the supervised bridge helper', async () => {
+    const registry = createRegistry()
+    const child = fakeChild()
+    const bwrapIdentity = identity()
+    const helperIdentity = {
+      ...bwrapIdentity,
+      executablePath: '/agent/node'
+    }
+    const spawn: DirectLinuxStdioSpawn = () => {
+      queueMicrotask(() => child.emit('spawn'))
+      return child
+    }
+    const readProcessIdentity = vi
+      .fn()
+      .mockResolvedValueOnce(bwrapIdentity)
+      .mockResolvedValue(helperIdentity)
+
+    const owner = await launchDirectLinuxStdioProcessOwner({
+      manifest: manifest(),
+      profile: {
+        ...profile(),
+        processExecutable: '/agent/node'
+      },
+      identity: { launchId: 'launch-1', processId: 'process-1' },
+      installationId: 'installation-1',
+      registry,
+      deadlineAt: '2030-01-01T00:00:00.000Z',
+      maximumInputBytes: 1024,
+      platform: 'linux',
+      spawn,
+      randomOwnerToken: () => 'a'.repeat(32),
+      readProcessIdentity
+    })
+
+    expect(readProcessIdentity).toHaveBeenCalledTimes(2)
+    expect(owner.processIdentity).toEqual(helperIdentity)
+    child.exitCode = 0
+    child.emit('close', 0, null)
+    registry.close()
+  })
+
   it('rejects PID reuse before sending any signal', async () => {
     const registry = createRegistry()
     const child = fakeChild()
@@ -208,6 +252,48 @@ describe('direct Linux stdio Runtime ownership', () => {
     registry.close()
   })
 
+  it('does not stop an active prompt at the unbounded Runtime deadline', async () => {
+    vi.useFakeTimers()
+    try {
+      const registry = createRegistry()
+      const child = fakeChild()
+      const launched = identity()
+      const sendSignal = vi.fn<typeof process.kill>(() => true)
+      const owner = await launchDirectLinuxStdioProcessOwner({
+        manifest: manifest(),
+        profile: profile(),
+        identity: { launchId: 'launch-long', processId: 'process-long' },
+        installationId: 'installation-1',
+        registry,
+        deadlineAt: UNBOUNDED_REMOTE_PROMPT_DEADLINE,
+        maximumInputBytes: 1024,
+        platform: 'linux',
+        spawn: () => {
+          queueMicrotask(() => child.emit('spawn'))
+          return child
+        },
+        randomOwnerToken: () => 'a'.repeat(32),
+        readProcessIdentity: async () => launched,
+        listPidNamespaceMembers: async () => [],
+        sendSignal
+      })
+
+      await vi.advanceTimersByTimeAsync(25 * 60 * 60_000)
+      expect(sendSignal).not.toHaveBeenCalled()
+      expect(await owner.reconcile()).toMatchObject({
+        state: 'running',
+        processTree: 'running'
+      })
+      await owner.stop({
+        reason: 'user-cancelled',
+        deadlineAt: new Date(Date.now() + 10_000).toISOString()
+      })
+      registry.close()
+    } finally {
+      vi.useRealTimers()
+    }
+  })
+
   it('startup reconciles only owners from the current installation', async () => {
     const registry = createRegistry()
     const reservation = {

src/agent-daemon/direct-linux-stdio-process-owner.ts

@@ -8,6 +8,9 @@ import {
   sha256DigestSchema
 } from '../shared/agent-protocol/contracts'
 import { canonicalJson } from '../shared/agent-protocol/canonical'
+import {
+  UNBOUNDED_REMOTE_PROMPT_DEADLINE
+} from '../shared/remote-agent-contracts'
 import {
   remoteRuntimeBundleManifestSchema,
   type RemoteRuntimeBundleManifest
@@ -31,6 +34,7 @@ const DEFAULT_MAXIMUM_STDIN_WRITE_BYTES = 1024 * 1024
 const DEFAULT_MAXIMUM_OUTPUT_QUEUE_CHUNKS = 128
 const DEFAULT_MAXIMUM_LISTENERS = 8
 const DEFAULT_STOP_TIMEOUT_MS = 10_000
+const DEFAULT_LAUNCH_IDENTITY_TIMEOUT_MS = 5_000
 const TERM_GRACE_MS = 2_000
 const POLL_INTERVAL_MS = 25
 const OWNER_TOKEN_ENVIRONMENT_NAME = 'GOODBUDDY_RUNTIME_OWNER_TOKEN'
@@ -510,6 +514,10 @@ export class DirectLinuxStdioProcessOwner {
   }
 
   #scheduleDeadline(deadlineAt: string): void {
+    if (deadlineAt === UNBOUNDED_REMOTE_PROMPT_DEADLINE) {
+      this.#clearDeadline()
+      return
+    }
     const remaining = remainingMilliseconds(deadlineAt, this.#now())
     this.#clearDeadline()
     const generation = this.#promptGeneration
@@ -550,7 +558,10 @@ export async function launchDirectLinuxStdioProcessOwner(
     now(),
     limits.maximumPromptRuntimeMilliseconds
   )
-  const timeoutMs = remainingMilliseconds(effectiveDeadlineAt, now())
+  const timeoutMs =
+    effectiveDeadlineAt === UNBOUNDED_REMOTE_PROMPT_DEADLINE
+      ? DEFAULT_LAUNCH_IDENTITY_TIMEOUT_MS
+      : remainingMilliseconds(effectiveDeadlineAt, now())
   const baseProfile = resolveProfile(options, manifest)
   const ownerToken = (options.randomOwnerToken ?? (() => randomBytes(16).toString('hex')))()
   if (!/^[a-f0-9]{32}$/u.test(ownerToken)) {
@@ -603,8 +614,18 @@ export async function launchDirectLinuxStdioProcessOwner(
     await waitForSpawn(child, options.signal, timeoutMs)
     if (child.pid === undefined) throw new Error('Runtime spawn did not return a PID')
     const readIdentity = options.readProcessIdentity ?? readLinuxRuntimeProcessIdentity
-    const processIdentity = await readIdentity(child.pid)
+    const processIdentity = await waitForLaunchedIdentity({
-    assertLaunchedIdentity(processIdentity, profile)
+      child,
+      pid: child.pid,
+      profile,
+      readIdentity,
+      signal: options.signal,
+      deadline: Math.min(
+        Date.parse(effectiveDeadlineAt),
+        now() + DEFAULT_LAUNCH_IDENTITY_TIMEOUT_MS
+      ),
+      now
+    })
     const supervisorIdentityDigest = supervisorDigest({
       installationId,
       ownerId,
@@ -871,15 +892,72 @@ function validateProfile(profile: OpenCodeLaunchProfile): OpenCodeLaunchProfile
   }
 }
 
-function assertLaunchedIdentity(
+async function waitForLaunchedIdentity(input: {
-  identity: LinuxRuntimeProcessIdentity,
+  child: DirectLinuxStdioChild
+  pid: number
   profile: OpenCodeLaunchProfile
-): void {
+  readIdentity: typeof readLinuxRuntimeProcessIdentity
-  if (
+  signal?: AbortSignal
-    identity.executablePath !== profile.processExecutable ||
+  deadline: number
-    identity.processGroupId !== identity.pid
+  now: () => number
-  ) {
+}): Promise<LinuxRuntimeProcessIdentity> {
-    throw identityError('Spawned Runtime does not match the launch identity')
+  for (;;) {
+    input.signal?.throwIfAborted()
+    if (
+      input.child.exitCode !== null ||
+      input.child.signalCode !== null
+    ) {
+      throw new DirectLinuxStdioProcessOwnerError(
+        'Runtime process exited during launch',
+        'process'
+      )
+    }
+    let identity: LinuxRuntimeProcessIdentity
+    try {
+      identity = await input.readIdentity(input.pid)
+    } catch (error) {
+      if (!isMissingProcess(error)) {
+        throw error
+      }
+      if (input.now() >= input.deadline) {
+        throw new DirectLinuxStdioProcessOwnerError(
+          'Runtime launch identity was not observed',
+          'process'
+        )
+      }
+      await delay(
+        Math.min(
+          POLL_INTERVAL_MS,
+          Math.max(1, input.deadline - input.now())
+        )
+      )
+      continue
+    }
+    if (identity.processGroupId !== identity.pid) {
+      throw identityError(
+        'Spawned Runtime does not match the launch identity'
+      )
+    }
+    if (identity.executablePath === input.profile.processExecutable) {
+      return identity
+    }
+    if (identity.executablePath !== input.profile.executable) {
+      throw identityError(
+        'Spawned Runtime does not match the launch identity'
+      )
+    }
+    if (input.now() >= input.deadline) {
+      throw new DirectLinuxStdioProcessOwnerError(
+        'Runtime launch did not reach its supervised executable',
+        'process'
+      )
+    }
+    await delay(
+      Math.min(
+        POLL_INTERVAL_MS,
+        Math.max(1, input.deadline - input.now())
+      )
+    )
   }
 }
 
@@ -1029,6 +1107,9 @@ function boundedPromptDeadline(
   now: number,
   maximumRuntimeMilliseconds: number
 ): string {
+  if (requestedDeadlineAt === UNBOUNDED_REMOTE_PROMPT_DEADLINE) {
+    return requestedDeadlineAt
+  }
   const requested = Date.parse(requestedDeadlineAt)
   if (!Number.isFinite(requested) || requested <= now) {
     throw new DirectLinuxStdioProcessOwnerError(

src/agent-daemon/opencode-runtime-profile.test.ts

@@ -241,6 +241,9 @@ describe('OpenCode bubblewrap launch profile', () => {
       '--opencode-entrypoint',
       resolve(fixture.bundleDirectory, 'bin', 'opencode')
     ])
+    expect(profile.processExecutable).toBe(
+      resolve(agentInstallationDirectory, 'node')
+    )
     expect(profile.args.join('\0')).not.toMatch(
       /api[-_]?key|authorization|bearer|(?:^|\0)--token(?:\0|$)/iu
     )

src/agent-daemon/opencode-runtime-profile.ts

@@ -266,7 +266,7 @@ export function createOpenCodeLaunchProfile(input: {
 
   return {
     executable: OPENCODE_BWRAP_EXECUTABLE,
-    processExecutable: OPENCODE_BWRAP_EXECUTABLE,
+    processExecutable: runtimeCommand.executable,
     args,
     cwd: scratchDirectory,
     env: {

src/agent-daemon/runtime-acp-backend.test.ts

@@ -1,7 +1,8 @@
 import { describe, expect, it, vi } from 'vitest'
 import {
   type AcpJournalCursor,
-  type RemotePromptOperationPreparation
+  type RemotePromptOperationPreparation,
+  UNBOUNDED_REMOTE_PROMPT_DEADLINE
 } from '../shared/remote-agent-contracts'
 import type { RemoteRuntimeBundleManifest } from '../shared/remote-runtime-launch-contracts'
 import type { ControllerLease } from './controller-registry'
@@ -813,6 +814,28 @@ describe('RuntimeAcpBackend', () => {
     ).rejects.toMatchObject({ code: 'deadline' })
   })
 
+  it('keeps an unbounded Runtime prompt alive without a wall-clock timer', async () => {
+    vi.useFakeTimers()
+    try {
+      const fixture = harness()
+      await open(fixture)
+      await invoke(
+        fixture,
+        'runtime/preparePrompt',
+        fixture.preparation({
+          deadlineAt: UNBOUNDED_REMOTE_PROMPT_DEADLINE
+        })
+      )
+
+      await vi.advanceTimersByTimeAsync(30 * 24 * 60 * 60_000)
+      expect(fixture.process.stops).toEqual([])
+      await fixture.process.emit('stdout', 'still-running')
+      expect(fixture.outputFrames).toEqual(['still-running'])
+    } finally {
+      vi.useRealTimers()
+    }
+  })
+
   it('stops the process when its output quota is exceeded', async () => {
     const fixture = harness({ maximumOutputBytes: 3 })
     await open(fixture)

src/agent-daemon/runtime-acp-backend.ts

@@ -25,6 +25,7 @@ import {
   acpJournalCursorSchema,
   remotePromptOperationAcceptanceSchema,
   remotePromptOperationPreparationSchema,
+  UNBOUNDED_REMOTE_PROMPT_DEADLINE,
   type AcpJournalCursor,
   type RemotePromptOperationAcceptance,
   type RemotePromptOperationPreparation
@@ -1676,6 +1677,10 @@ export class RuntimeAcpBackend {
   #scheduleDeadline(binding: BindingState, deadlineAt: string): void {
     if (binding.deadlineTimer !== undefined) {
       clearTimeout(binding.deadlineTimer)
+      binding.deadlineTimer = undefined
+    }
+    if (deadlineAt === UNBOUNDED_REMOTE_PROMPT_DEADLINE) {
+      return
     }
     const delay = Math.max(
       1,

src/main/agent/acp-remote-runtime.test.ts

@@ -259,10 +259,9 @@ function runtime(
     assertHostCurrent: () => {},
     cancellationGraceMs,
     operationTimeoutMs: overrides.operationTimeoutMs ?? 100,
-    promptTimeoutMs:
+    ...(overrides.promptTimeoutMs === undefined
-      overrides.promptTimeoutMs ??
+      ? {}
-      overrides.operationTimeoutMs ??
+      : { promptTimeoutMs: overrides.promptTimeoutMs }),
-      100,
     usage: {
       runtime: 'opencode',
       provider: 'test',
@@ -287,10 +286,9 @@ function factoryRuntime(
     assertHostCurrent: () => {},
     cancellationGraceMs: 5,
     operationTimeoutMs: overrides.operationTimeoutMs ?? 100,
-    promptTimeoutMs:
+    ...(overrides.promptTimeoutMs === undefined
-      overrides.promptTimeoutMs ??
+      ? {}
-      overrides.operationTimeoutMs ??
+      : { promptTimeoutMs: overrides.promptTimeoutMs }),
-      100,
     usage: {
       runtime: 'opencode',
       provider: 'test',
@@ -599,7 +597,7 @@ describe('AcpRemoteRuntime', () => {
   })
 
   it.each(['prompt-running', 'outcome-unknown'] as const)(
-    'does not replace a %s binding after identity changes',
+    'retires a %s binding before starting an explicit later prompt',
     async (state) => {
       const store = new MemoryRuntimeSessionBindingStore()
       const active = {
@@ -625,7 +623,13 @@ describe('AcpRemoteRuntime', () => {
         lastMainAckSequence: '0'
       }
       await store.put(active)
-      const factory = vi.fn(async () => fakeServer().channel)
+      const replacement = fakeServer()
+      const factory = vi.fn(
+        async (bindingId: string) => {
+          expect(bindingId).not.toBe(active.bindingId)
+          return replacement.channel
+        }
+      )
       const instance = factoryRuntime(factory, store, {
         identity: {
           ...identity,
@@ -633,22 +637,28 @@ describe('AcpRemoteRuntime', () => {
         }
       })
 
-      await expect(
+      await collect(
-        collect(
+        instance.run(
-          instance.run(
+          { ...request, requestId: 'prompt-2' },
-            { ...request, requestId: 'prompt-2' },
+          new AbortController().signal
-            new AbortController().signal
-          )
         )
-      ).rejects.toThrow('不会自动重放')
+      )
 
-      expect(factory).not.toHaveBeenCalled()
+      expect(factory).toHaveBeenCalledOnce()
+      expect(factory).not.toHaveBeenCalledWith(active.bindingId)
+      await expect(
+        store.getById(active.bindingId)
+      ).resolves.toMatchObject({
+        bindingId: active.bindingId,
+        state: 'closed',
+        activePromptOperationId: undefined
+      })
       await expect(
         store.getByConversation(request.conversationId)
       ).resolves.toMatchObject({
-        bindingId: active.bindingId,
+        bindingId: factory.mock.calls[0]![0],
-        state: 'outcome-unknown',
+        state: 'ready',
-        activePromptOperationId: request.requestId
+        activePromptOperationId: undefined
       })
       await instance.dispose()
     }
@@ -1246,6 +1256,233 @@ describe('AcpRemoteRuntime', () => {
     ])
   })
 
+  it('merges incremental OpenCode Task input into native subagent events', async () => {
+    const client: { value?: AgentSideConnection } = {}
+    const server = fakeServer({
+      prompt: async ({ sessionId }) => {
+        await client.value!.sessionUpdate({
+          sessionId,
+          update: {
+            sessionUpdate: 'tool_call',
+            toolCallId: 'task-1',
+            title: 'Review application architecture',
+            kind: 'other',
+            status: 'in_progress',
+            rawInput: {}
+          }
+        })
+        await client.value!.sessionUpdate({
+          sessionId,
+          update: {
+            sessionUpdate: 'tool_call_update',
+            toolCallId: 'task-1',
+            status: 'in_progress',
+            rawInput: {
+              subagent_type: 'explorer',
+              description: 'Review application architecture',
+              prompt: 'Inspect the complete source tree.'
+            }
+          }
+        })
+        await client.value!.sessionUpdate({
+          sessionId,
+          update: {
+            sessionUpdate: 'tool_call_update',
+            toolCallId: 'task-1',
+            status: 'in_progress',
+            rawInput: {
+              subagent_type: 'explorer',
+              description: 'Review application architecture',
+              prompt: 'Inspect the complete source tree.'
+            }
+          }
+        })
+        await client.value!.sessionUpdate({
+          sessionId,
+          update: {
+            sessionUpdate: 'tool_call_update',
+            toolCallId: 'task-1',
+            status: 'completed',
+            rawOutput: 'Architecture review complete.'
+          }
+        })
+        return { stopReason: 'end_turn' }
+      }
+    })
+    client.value = server.client
+
+    const events = await collect(
+      runtime(server).run(
+        request,
+        new AbortController().signal,
+        async () => 'once'
+      )
+    )
+
+    expect(events.map((event) => event.type)).toEqual([
+      'status',
+      'tool',
+      'subagent',
+      'subagent',
+      'done'
+    ])
+    expect(events[2]).toMatchObject({
+      expertName: 'explorer',
+      routingMode: 'native',
+      runtimeCallId: 'task-1',
+      state: 'running'
+    })
+    expect(events[3]).toMatchObject({
+      state: 'completed',
+      output: 'Architecture review complete.'
+    })
+  })
+
+  it('bounds native subagent output per event and per request', async () => {
+    const client: { value?: AgentSideConnection } = {}
+    const server = fakeServer({
+      prompt: async ({ sessionId }) => {
+        await client.value!.sessionUpdate({
+          sessionId,
+          update: {
+            sessionUpdate: 'tool_call',
+            toolCallId: 'task-bounded',
+            title: 'Bounded task',
+            kind: 'other',
+            status: 'completed',
+            rawInput: {
+              subagent_type: 'explorer',
+              prompt: 'Inspect the complete source tree.'
+            },
+            rawOutput: 'x'.repeat(100)
+          }
+        })
+        return { stopReason: 'end_turn' }
+      }
+    })
+    client.value = server.client
+
+    const events = await collect(
+      runtime(server, undefined, 5, {
+        maxEventCharacters: 10,
+        maxRequestOutputBytes: 1_000
+      }).run(
+        request,
+        new AbortController().signal,
+        async () => 'once'
+      )
+    )
+    expect(
+      events.find((event) => event.type === 'subagent')
+    ).toMatchObject({
+      state: 'completed',
+      output: 'xxxxx…'
+    })
+
+    const overBudgetServer = fakeServer({
+      prompt: async ({ sessionId }) => {
+        await client.value!.sessionUpdate({
+          sessionId,
+          update: {
+            sessionUpdate: 'tool_call',
+            toolCallId: 'task-over-budget',
+            title: 'Over-budget task',
+            kind: 'other',
+            status: 'completed',
+            rawInput: {
+              subagent_type: 'explorer',
+              prompt: 'Inspect the complete source tree.'
+            },
+            rawOutput: 'x'.repeat(100)
+          }
+        })
+        return { stopReason: 'end_turn' }
+      }
+    })
+    client.value = overBudgetServer.client
+    await expect(
+      collect(
+        runtime(overBudgetServer, undefined, 5, {
+          maxEventCharacters: 1_000,
+          maxRequestOutputBytes: 80
+        }).run(
+          {
+            ...request,
+            requestId: 'b5b07fa5-f722-4bbc-ad19-33a6435817f1'
+          },
+          new AbortController().signal,
+          async () => 'once'
+        )
+      )
+    ).rejects.toThrow('远端 Runtime 请求输出超过安全上限')
+  })
+
+  it('rejects more than 100 distinct remote tool calls', async () => {
+    const client: { value?: AgentSideConnection } = {}
+    const server = fakeServer({
+      prompt: async ({ sessionId }) => {
+        for (let index = 0; index <= 100; index += 1) {
+          await client.value!.sessionUpdate({
+            sessionId,
+            update: {
+              sessionUpdate: 'tool_call',
+              toolCallId: `call-${index}`,
+              title: 'tool',
+              status: 'in_progress'
+            }
+          })
+        }
+        return { stopReason: 'end_turn' }
+      }
+    })
+    client.value = server.client
+
+    await expect(
+      collect(
+        runtime(server).run(
+          request,
+          new AbortController().signal,
+          async () => 'once'
+        )
+      )
+    ).rejects.toThrow(
+      '远端 Runtime 单次运行的工具调用超过 100 个'
+    )
+  })
+
+  it('does not impose a default wall-clock deadline on remote prompts', async () => {
+    vi.useFakeTimers()
+    try {
+      let release!: () => void
+      const gate = new Promise<void>((resolve) => {
+        release = resolve
+      })
+      const server = fakeServer({
+        prompt: async () => {
+          await gate
+          return { stopReason: 'end_turn' }
+        }
+      })
+      const running = collect(
+        runtime(server).run(
+          request,
+          new AbortController().signal,
+          async () => 'once'
+        )
+      )
+      await vi.waitFor(() => expect(server.prepare).toHaveBeenCalledOnce())
+      await vi.advanceTimersByTimeAsync(25 * 60 * 60_000)
+      expect(server.escalate).not.toHaveBeenCalled()
+      release()
+      await expect(running).resolves.toEqual([
+        expect.objectContaining({ type: 'status' }),
+        expect.objectContaining({ type: 'done' })
+      ])
+    } finally {
+      vi.useRealTimers()
+    }
+  })
+
   it('routes Execute permission through the authorizer and denies mutating Ask permission', async () => {
     const client: { value?: AgentSideConnection } = {}
     const outcomes: string[] = []
@@ -1715,7 +1952,8 @@ describe('AcpRemoteRuntime', () => {
       })
     })
     const instance = runtime(server, store, 5, {
-      operationTimeoutMs: 20
+      operationTimeoutMs: 20,
+      promptTimeoutMs: 20
     })
 
     await expect(

src/main/agent/acp-remote-runtime.ts

@@ -15,11 +15,15 @@ import {
 import type {
   AgentRuntimeStatus
 } from '../../shared/contracts'
+import {
+  maximumConversationToolActivities
+} from '../../shared/assistant-contracts'
 import {
   assertRemotePromptAcceptanceMatchesPreparation,
   remotePromptOperationAcceptanceSchema,
   remotePromptOperationPreparationSchema,
   runtimeSessionBindingSchema,
+  UNBOUNDED_REMOTE_PROMPT_DEADLINE,
   type RemotePromptOperationAcceptance,
   type RemotePromptOperationPreparation,
   type RuntimeSessionBinding
@@ -48,6 +52,10 @@ import {
   type RemoteRuntimeChannel
 } from './remote-runtime-channel'
 import type { RuntimeSessionBindingStore } from './runtime-session-binding-store'
+import {
+  parseOpenCodeSubagentInput,
+  toOpenCodeSubagentEvent
+} from './opencode-subagent'
 
 type BindingIdentity = Pick<
   RuntimeSessionBinding,
@@ -98,6 +106,7 @@ export type AcpRemoteRuntimeOptions = {
   maxRequestOutputBytes?: number
   cancellationGraceMs?: number
   operationTimeoutMs?: number
+  /** Optional test/operator override. Production prompts are Runtime-controlled. */
   promptTimeoutMs?: number
   maxPendingUpdates?: number
   maxPendingUpdateBytes?: number
@@ -120,6 +129,16 @@ type ActivePrompt = {
   pendingUpdateBytes: number
   inboundPaused: boolean
   outputCharacters: number
+  toolCalls: Map<
+    string,
+    {
+      title?: string
+      kind?: string
+      status?: 'pending' | 'in_progress' | 'completed' | 'failed'
+      rawInput?: unknown
+      rawOutput?: unknown
+    }
+  >
   open: boolean
   completed: boolean
   interruption?: Error
@@ -162,7 +181,6 @@ type ChannelContext = {
 const DEFAULT_MAX_EVENT_CHARACTERS = 100_000
 const DEFAULT_MAX_REQUEST_OUTPUT_BYTES = 1_000_000
 const DEFAULT_OPERATION_TIMEOUT_MS = 30_000
-const DEFAULT_PROMPT_TIMEOUT_MS = 120_000
 const DEFAULT_CANCELLATION_GRACE_MS = 1_500
 const DEFAULT_MAX_PENDING_UPDATES = 1_000
 const DEFAULT_MAX_PENDING_UPDATE_BYTES = 4 * 1024 * 1024
@@ -188,6 +206,42 @@ function safeStringify(value: unknown, maximum: number): string | undefined {
   }
 }
 
+function boundedRetainedValue(
+  value: unknown,
+  maximum: number
+): unknown {
+  if (value === undefined) {
+    return undefined
+  }
+  if (typeof value === 'string') {
+    return value.length > maximum
+      ? `${value.slice(0, maximum)}…`
+      : value
+  }
+  try {
+    const serialized = JSON.stringify(value)
+    if (serialized === undefined) {
+      return undefined
+    }
+    if (serialized.length > maximum) {
+      return `${serialized.slice(0, maximum)}…`
+    }
+    return JSON.parse(serialized) as unknown
+  } catch {
+    return undefined
+  }
+}
+
+function boundedRetainedToolInput(
+  value: unknown,
+  maximum: number
+): unknown {
+  return (
+    parseOpenCodeSubagentInput(value) ??
+    boundedRetainedValue(value, maximum)
+  )
+}
+
 class RemoteOperationTimeoutError extends Error {
   constructor(readonly operation: string) {
     super(`远端 Runtime ${operation}超时`)
@@ -459,9 +513,8 @@ export class AcpRemoteRuntime implements AgentRuntime {
       DEFAULT_OPERATION_TIMEOUT_MS
   }
 
-  private get promptTimeoutMs(): number {
+  private get promptTimeoutMs(): number | undefined {
-    return this.options.promptTimeoutMs ??
+    return this.options.promptTimeoutMs
-      DEFAULT_PROMPT_TIMEOUT_MS
   }
 
   private async awaitOperation<T>(
@@ -833,7 +886,7 @@ export class AcpRemoteRuntime implements AgentRuntime {
     return interrupted
   }
 
-  private async persistRejectedPrompt(
+  private async persistClosedBinding(
     binding: RuntimeSessionBinding
   ): Promise<RuntimeSessionBinding> {
     const closed = runtimeSessionBindingSchema.parse({
@@ -982,9 +1035,10 @@ export class AcpRemoteRuntime implements AgentRuntime {
     modelBridge?: RemoteModelBridgeSession
   ): RemotePromptOperationPreparation {
     const workMode = request.workMode === 'execute' ? 'execute' : 'ask'
-    const deadlineAt = new Date(
+    const deadlineAt =
-      Date.now() + this.promptTimeoutMs
+      this.promptTimeoutMs === undefined
-    ).toISOString()
+        ? UNBOUNDED_REMOTE_PROMPT_DEADLINE
+        : new Date(Date.now() + this.promptTimeoutMs).toISOString()
     const budget = {
       // The Agent applies this to the complete ACP stdin stream, not only
       // user content. The signed Runtime manifest supplies the effective
@@ -1198,24 +1252,19 @@ export class AcpRemoteRuntime implements AgentRuntime {
     request: AgentExecutionRequest,
     signal: AbortSignal
   ): Promise<SessionRecord> {
-    const old = await this.awaitLocalOperation(
+    let binding = await this.awaitLocalOperation(
       '读取会话绑定',
       this.options.bindingStore.getByConversation(conversationId),
       signal
     )
     if (
-      old &&
+      binding &&
-      old.state !== 'closed' &&
+      binding.state !== 'closed' &&
-      old.state !== 'ready'
+      binding.state !== 'ready'
     ) {
-      await this.persistInterrupted(old)
+      await this.persistClosedBinding(binding)
-      throw new Error(
+      binding = undefined
-        old.state === 'prompt-running'
-          ? '远端 ACP 会话存在未确认的活动请求，结果未知且不会自动重放'
-          : '远端 ACP 会话恢复状态不确定，已中断且不会自动重放'
-      )
     }
-    let binding = old
     let bindingId = binding?.bindingId ?? randomUUID()
     if (
       binding?.state === 'ready' &&
@@ -1437,7 +1486,7 @@ export class AcpRemoteRuntime implements AgentRuntime {
             error,
             'runtime/preparePrompt'
           )
-            ? this.persistRejectedPrompt(binding)
+            ? this.persistClosedBinding(binding)
             : this.persistInterrupted(binding)
         await persist.catch(() => undefined)
       }
@@ -1483,7 +1532,7 @@ export class AcpRemoteRuntime implements AgentRuntime {
             error,
             'runtime/preparePrompt'
           )
-            ? this.persistRejectedPrompt(current.binding)
+            ? this.persistClosedBinding(current.binding)
             : this.persistInterrupted(current.binding)
         current.binding = await persist.catch(() => current.binding)
         this.sessions.delete(conversationId)
@@ -1776,6 +1825,21 @@ export class AcpRemoteRuntime implements AgentRuntime {
         (event.input?.length ?? 0) +
         (event.output?.length ?? 0) +
         (event.error?.length ?? 0)
+    } else if (event.type === 'subagent') {
+      const fieldMaximum = Math.max(1, Math.floor(maximum / 2))
+      const bounded = (value: string | undefined): string | undefined =>
+        value && value.length > fieldMaximum
+          ? `${value.slice(0, fieldMaximum)}…`
+          : value
+      event = {
+        ...event,
+        output: bounded(event.output),
+        error: bounded(event.error)
+      }
+      characters =
+        (event.reason?.length ?? 0) +
+        (event.output?.length ?? 0) +
+        (event.error?.length ?? 0)
     } else if (event.type === 'status') {
       if (event.message.length > maximum) {
         event = {
@@ -1819,22 +1883,87 @@ export class AcpRemoteRuntime implements AgentRuntime {
       update.sessionUpdate === 'tool_call' ||
       update.sessionUpdate === 'tool_call_update'
     ) {
+      const callId = update.toolCallId.slice(0, 256)
+      const previous = prompt.toolCalls.get(callId)
+      if (
+        !previous &&
+        prompt.toolCalls.size >=
+          maximumConversationToolActivities
+      ) {
+        throw new Error(
+          '远端 Runtime 单次运行的工具调用超过 100 个'
+        )
+      }
+      const retainedMaximum =
+        this.options.maxEventCharacters ??
+        DEFAULT_MAX_EVENT_CHARACTERS
+      const toolCall = {
+        title:
+          update.title === null
+            ? undefined
+            : update.title?.slice(0, 200) ?? previous?.title,
+        kind:
+          update.kind === null
+            ? undefined
+            : update.kind?.slice(0, 200) ?? previous?.kind,
+        status:
+          update.status === null
+            ? undefined
+            : update.status ?? previous?.status,
+        rawInput:
+          update.rawInput === undefined
+            ? previous?.rawInput
+            : boundedRetainedToolInput(
+                update.rawInput,
+                retainedMaximum
+              ),
+        rawOutput:
+          update.rawOutput === undefined
+            ? previous?.rawOutput
+            : boundedRetainedValue(
+                update.rawOutput,
+                retainedMaximum
+              )
+      }
+      if (previous && isDeepStrictEqual(previous, toolCall)) {
+        return undefined
+      }
+      prompt.toolCalls.set(callId, toolCall)
       const name = (
-        update.title ??
+        toolCall.title ??
-        update.kind ??
+        toolCall.kind ??
         '远端 Runtime 工具'
       ).slice(0, 200)
-      const input = safeStringify(update.rawInput, 100_000)
+      const subagent =
-      const output = safeStringify(update.rawOutput, 100_000)
+        this.options.runtimeId === 'opencode'
+          ? toOpenCodeSubagentEvent({
+              requestId: prompt.requestId,
+              callId,
+              state: toolCall.status ?? 'pending',
+              input: toolCall.rawInput,
+              output: toolCall.rawOutput
+            })
+          : undefined
+      if (subagent) {
+        return subagent
+      }
+      const input = safeStringify(
+        toolCall.rawInput,
+        retainedMaximum
+      )
+      const output = safeStringify(
+        toolCall.rawOutput,
+        retainedMaximum
+      )
       return {
         requestId: prompt.requestId,
         type: 'tool',
-        callId: update.toolCallId.slice(0, 256),
+        callId,
         name,
         state:
-          update.status === 'in_progress'
+          toolCall.status === 'in_progress'
             ? 'running'
-            : update.status ?? 'pending',
+            : toolCall.status ?? 'pending',
         summary: `远端 Runtime 工具：${name}`,
         ...(input ? { input } : {}),
         ...(output ? { output } : {})
@@ -1930,6 +2059,7 @@ export class AcpRemoteRuntime implements AgentRuntime {
       pendingUpdateBytes: 0,
       inboundPaused: false,
       outputCharacters: 0,
+      toolCalls: new Map(),
       open: true,
       completed: false,
       context
@@ -1998,10 +2128,12 @@ export class AcpRemoteRuntime implements AgentRuntime {
           prompt.wake?.()
           prompt.wake = undefined
         })
-      promptTimeout = setTimeout(
+      if (this.promptTimeoutMs !== undefined) {
-        () => interrupt(new RemoteOperationTimeoutError('执行请求')),
+        promptTimeout = setTimeout(
-        this.promptTimeoutMs
+          () => interrupt(new RemoteOperationTimeoutError('执行请求')),
-      )
+          this.promptTimeoutMs
+        )
+      }
       yield {
         requestId: request.requestId,
         type: 'status',

src/main/agent/continue-host-adapter.test.ts

@@ -357,6 +357,9 @@ describe('ContinueHostAdapter', () => {
   })
 
   it('launches the prepared host through the injected launcher', async () => {
+    let now = Date.now()
+    const nowSpy = vi.spyOn(Date, 'now').mockImplementation(() => now)
+    environmentRestorations.push(() => nowSpy.mockRestore())
     const distribution = await createDistribution()
     const skillDirectory = join(
       distribution.cacheRoot,
@@ -445,11 +448,14 @@ describe('ContinueHostAdapter', () => {
         const url = String(input)
         if (url.endsWith('/state')) {
           stateRequests += 1
+          if (stateRequests === 2) {
+            now += 25 * 60 * 60_000
+          }
           return new Response(
             JSON.stringify({
               session: {
                 history:
-                  stateRequests === 1
+                  stateRequests <= 2
                     ? []
                     : [
                         {
@@ -460,7 +466,7 @@ describe('ContinueHostAdapter', () => {
                         }
                       ],
                 usage:
-                  stateRequests === 1
+                  stateRequests <= 2
                     ? {
                         promptTokens: 20,
                         completionTokens: 10,
@@ -478,7 +484,7 @@ describe('ContinueHostAdapter', () => {
                         }
                       }
               },
-              isProcessing: false,
+              isProcessing: stateRequests === 2,
               messageQueueLength: 0,
               pendingPermission: null
             })
@@ -605,6 +611,90 @@ describe('ContinueHostAdapter', () => {
     expect(existsSync(generatedConfigPath)).toBe(false)
   })
 
+  it('bounds each host control request without bounding the full run', async () => {
+    const distribution = await createDistribution()
+    let killed = false
+    const launchHost: ContinueHostLauncher = () => ({
+      exitCode: null,
+      get killed() {
+        return killed
+      },
+      stderr: null,
+      once: () => undefined,
+      kill: () => {
+        killed = true
+        return true
+      }
+    })
+    vi.stubGlobal(
+      'fetch',
+      vi.fn(
+        async (
+          input: string | URL | Request,
+          init?: RequestInit
+        ): Promise<Response> => {
+          if (String(input).endsWith('/state')) {
+            return Response.json({
+              session: { history: [] },
+              isProcessing: false,
+              messageQueueLength: 0,
+              pendingPermission: null
+            })
+          }
+          if (String(input).endsWith('/message')) {
+            return await new Promise<Response>((_resolve, reject) => {
+              const rejectOnAbort = (): void =>
+                reject(init?.signal?.reason)
+              init?.signal?.addEventListener(
+                'abort',
+                rejectOnAbort,
+                { once: true }
+              )
+              if (init?.signal?.aborted) {
+                rejectOnAbort()
+              }
+            })
+          }
+          return Response.json({})
+        }
+      )
+    )
+    const adapter = new ContinueHostAdapter(
+      {
+        binaryPath: distribution.entryPath,
+        configPath: '',
+        workspace: process.cwd(),
+        cacheRoot: distribution.cacheRoot,
+        trustedBundleHashes: [distribution.sourceHash],
+        launchHost,
+        modelProfile: {
+          id: '00000000-0000-4000-8000-000000000012',
+          name: '独立模型',
+          baseUrl: 'https://model.example',
+          modelName: 'private-model',
+          protocol: 'anthropic-messages',
+          authentication: 'api-key',
+          apiKey: 'private-key'
+        }
+      },
+      {
+        controlRequestTimeoutMs: 5,
+        terminateProcessTree: vi.fn(async (child) => {
+          child.kill()
+        })
+      }
+    )
+
+    await expect(
+      adapter.run(
+        'hello',
+        new AbortController().signal,
+        async () => 'deny'
+      )
+    ).rejects.toThrow('Continue 宿主请求超时（/message）')
+    expect(killed).toBe(true)
+  })
+
   it('injects scoped knowledge into a temporary copy of a JSONC config', async () => {
     const distribution = await createDistribution()
     const configPath = join(

src/main/agent/continue-host-adapter.ts

@@ -26,6 +26,9 @@ import type {
   RuntimeNativeSnapshot,
   RuntimeSettings
 } from '../../shared/contracts'
+import {
+  maximumConversationToolActivities
+} from '../../shared/assistant-contracts'
 import {
   continueConfigurationPresetSchema,
   runtimeNativeInventoryLimits,
@@ -66,8 +69,8 @@ const maximumConfiguredRules = runtimeNativeInventoryLimits.rules
 const maximumConfiguredPrompts = runtimeNativeInventoryLimits.prompts
 const maximumStreamEvents = 5_000
 const maximumStreamEventBytes = 2 * 1024 * 1024
-const maximumToolCalls = 100
+const maximumToolCalls = maximumConversationToolActivities
-const maximumExecutionMilliseconds = 10 * 60_000
+const defaultControlRequestTimeoutMs = 30_000
 const knowledgeMcpName = 'goodbuddy-knowledge'
 const customMcpName = 'goodbuddy-custom-mcp'
 export const continueConfigurationRequiredMessage =
@@ -246,6 +249,7 @@ export type ContinueHostAdapterDependencies = {
   maximumStreamEvents: number
   maximumStreamEventBytes: number
   maximumToolCalls: number
+  controlRequestTimeoutMs: number
 }
 
 export type ContinueHostRunOptions = {
@@ -830,6 +834,7 @@ export class ContinueHostAdapter {
       maximumStreamEvents,
       maximumStreamEventBytes,
       maximumToolCalls,
+      controlRequestTimeoutMs: defaultControlRequestTimeoutMs,
       ...dependencies
     }
   }
@@ -1140,16 +1145,32 @@ export class ContinueHostAdapter {
     path: string,
     init: RequestInit = {}
   ): Promise<unknown> {
-    const response = await fetch(`${origin}${path}`, {
+    const timeoutSignal = AbortSignal.timeout(
-      ...init,
+      Math.max(1, this.dependencies.controlRequestTimeoutMs)
-      headers: {
+    )
-        authorization: `Bearer ${token}`,
+    const requestSignal = init.signal
-        'content-type': 'application/json',
+      ? AbortSignal.any([init.signal, timeoutSignal])
-        ...init.headers
+      : timeoutSignal
-      },
+    let response: Response
-      redirect: 'error',
+    try {
-      signal: init.signal
+      response = await fetch(`${origin}${path}`, {
-    })
+        ...init,
+        headers: {
+          authorization: `Bearer ${token}`,
+          'content-type': 'application/json',
+          ...init.headers
+        },
+        redirect: 'error',
+        signal: requestSignal
+      })
+    } catch (error) {
+      if (timeoutSignal.aborted && !init.signal?.aborted) {
+        throw new Error(`Continue 宿主请求超时（${path}）`, {
+          cause: error
+        })
+      }
+      throw error
+    }
     const body = await readBoundedResponseText(response, {
       maxBytes: maximumStateBytes,
       tooLargeMessage: 'Continue 宿主响应超过安全大小限制'
@@ -1559,7 +1580,6 @@ export class ContinueHostAdapter {
     let streamEventCount = 0
     let streamEventBytes = 0
     const observedToolCallIds = new Set<string>()
-    let executionTimeoutSignal: AbortSignal | undefined
     try {
       const initialState = await this.waitForStartup(
         child,
@@ -1569,13 +1589,7 @@ export class ContinueHostAdapter {
         signal
       )
       const startIndex = initialState.session.history.length
-      executionTimeoutSignal = AbortSignal.timeout(
+      const executionSignal = signal
-        maximumExecutionMilliseconds
-      )
-      const executionSignal = AbortSignal.any([
-        signal,
-        executionTimeoutSignal
-      ])
       const message =
         runOptions.images && runOptions.images.length > 0
           ? [
@@ -1598,9 +1612,8 @@ export class ContinueHostAdapter {
         signal: executionSignal
       })
 
-      const expiresAt = Date.now() + maximumExecutionMilliseconds
       const handledPermissionIds = new Set<string>()
-      while (Date.now() < expiresAt) {
+      while (true) {
         executionSignal.throwIfAborted()
         if (childFailure) {
           throw childFailure
@@ -1817,20 +1830,15 @@ export class ContinueHostAdapter {
         }
         await delay(150, executionSignal)
       }
-      throw new Error('Continue 宿主执行超时')
     } catch (error) {
       if (error instanceof ContinueHostRunError) {
         throw error
       }
-      const normalizedError =
-        executionTimeoutSignal?.aborted && !signal.aborted
-          ? new Error('Continue 宿主执行超时', { cause: error })
-          : error
       throw new ContinueHostRunError(
-        normalizedError instanceof Error
+        error instanceof Error
-          ? normalizedError.message
+          ? error.message
           : 'Continue 宿主执行失败',
-        { cause: normalizedError, tools: observedTools }
+        { cause: error, tools: observedTools }
       )
     } finally {
       signal.removeEventListener('abort', abort)

src/main/agent/create-runtime.test.ts

@@ -100,7 +100,7 @@ describe('createAgentRuntime model compatibility', () => {
         { deepseekHarnessLauncher: vi.fn() }
       )
     ).toThrow(
-      'DeepSeek Harness 需要使用 API Key 的安全 OpenAI 兼容 Chat Completions 模型连接'
+      'DeepSeek Harness 需要使用 API Key 的 OpenAI 兼容 Chat Completions 模型连接'
     )
   })
 

src/main/agent/create-runtime.ts

@@ -190,7 +190,7 @@ export function createAgentRuntime(
     const profile = settings?.deepseekHarnessModelProfile
     if (!profile || !isDeepSeekHarnessModelProfile(profile)) {
       throw new Error(
-        'DeepSeek Harness 需要使用 API Key 的安全 OpenAI 兼容 Chat Completions 模型连接'
+        'DeepSeek Harness 需要使用 API Key 的 OpenAI 兼容 Chat Completions 模型连接'
       )
     }
     if (!profile.apiKey) {

src/main/agent/deepseek-harness-runtime.test.ts

@@ -44,6 +44,7 @@ function setup(
     nativeTools?: Array<Record<string, unknown>>
     toolsSupported?: boolean
     promptTimeoutMs?: number
+    useProductionPromptTimeout?: boolean
     maxEventCharacters?: number
     maxRequestOutputCharacters?: number
     supportsImageInput?: boolean
@@ -51,6 +52,7 @@ function setup(
     initializationTimeoutMs?: number
     useDefaultInitializationTimeout?: boolean
     launchDelayMs?: number
+    prepare?: () => Promise<Record<string, unknown>>
     extensionPackages?: Array<{
       id: string
       entrypoint: string
@@ -170,7 +172,7 @@ function setup(
           }
         }
         if (method === 'goodbuddy/session/prepare') {
-          return { prepared: true }
+          return options.prepare?.() ?? { prepared: true }
         }
         if (method === 'goodbuddy/session/release') {
           return { released: true }
@@ -253,7 +255,9 @@ function setup(
           initializationTimeoutMs:
             options.initializationTimeoutMs ?? 100
         }),
-    promptTimeoutMs: options.promptTimeoutMs ?? 100,
+    ...(options.useProductionPromptTimeout
+      ? {}
+      : { promptTimeoutMs: options.promptTimeoutMs ?? 100 }),
     shutdownTimeoutMs: 10,
     maxStderrBytes: 16,
     maxEventCharacters: options.maxEventCharacters,
@@ -1347,6 +1351,62 @@ describe('DeepSeekHarnessRuntime', () => {
     await harness.runtime.dispose()
   })
 
+  it('does not submit a prompt after cancellation during preparation', async () => {
+    const prepare = deferred<Record<string, unknown>>()
+    const harness = setup({ prepare: () => prepare.promise })
+    const controller = new AbortController()
+    const running = collect(
+      harness.runtime.run(
+        request('cancel-during-prepare'),
+        controller.signal
+      )
+    )
+    await vi.waitFor(() =>
+      expect(
+        harness.requests.some(
+          ({ method }) => method === 'goodbuddy/session/prepare'
+        )
+      ).toBe(true)
+    )
+
+    controller.abort(new Error('cancelled during preparation'))
+    prepare.resolve({ prepared: true })
+
+    await expect(running).rejects.toThrow(
+      'cancelled during preparation'
+    )
+    expect(harness.agent.prompt).not.toHaveBeenCalled()
+    expect(harness.notifications).toContainEqual({
+      method: 'session/cancel',
+      params: { sessionId: 'session-1' }
+    })
+    await harness.runtime.dispose()
+  })
+
+  it('bounds cancellation when the harness prompt never settles', async () => {
+    const harness = setup({ useProductionPromptTimeout: true })
+    const controller = new AbortController()
+    const running = collect(
+      harness.runtime.run(
+        request('uncooperative-cancel'),
+        controller.signal
+      )
+    )
+    await vi.waitFor(() =>
+      expect(harness.promptGates).toHaveLength(1)
+    )
+
+    controller.abort(new Error('cancelled by user'))
+
+    await expect(running).rejects.toThrow('cancelled by user')
+    expect(harness.notifications).toContainEqual({
+      method: 'session/cancel',
+      params: { sessionId: 'session-1' }
+    })
+    expect(harness.child.terminate).toHaveBeenCalled()
+    await harness.runtime.dispose()
+  })
+
   it('fails on bounded stderr overflow without exposing stderr text', async () => {
     const harness = setup()
     const running = collect(
@@ -1420,4 +1480,32 @@ describe('DeepSeekHarnessRuntime', () => {
     })
     await expect(harness.runtime.dispose()).resolves.toBeUndefined()
   })
+
+  it('does not impose a default wall-clock deadline on prompts', async () => {
+    vi.useFakeTimers()
+    try {
+      const harness = setup({ useProductionPromptTimeout: true })
+      const running = collect(
+        harness.runtime.run(
+          request('long-running'),
+          new AbortController().signal
+        )
+      )
+      await vi.waitFor(() =>
+        expect(harness.promptGates).toHaveLength(1)
+      )
+      await vi.advanceTimersByTimeAsync(25 * 60 * 60_000)
+      expect(harness.notifications).not.toContainEqual({
+        method: 'session/cancel',
+        params: { sessionId: 'session-1' }
+      })
+      harness.promptGates[0]!.resolve({ stopReason: 'end_turn' })
+      await expect(running).resolves.toEqual([
+        expect.objectContaining({ type: 'status' }),
+        expect.objectContaining({ type: 'done' })
+      ])
+    } finally {
+      vi.useRealTimers()
+    }
+  })
 })

src/main/agent/deepseek-harness-runtime.ts

@@ -38,7 +38,6 @@ import { deepSeekHarnessStartupBudget } from './deepseek-harness-control-protoco
 
 const ACP_PACKAGE_NAME = '@agentclientprotocol/sdk'
 const DEFAULT_INITIALIZATION_TIMEOUT_MS = 10_000
-const DEFAULT_PROMPT_TIMEOUT_MS = 10 * 60_000
 const DEFAULT_SHUTDOWN_TIMEOUT_MS = 2_000
 const DEFAULT_MAX_STDERR_BYTES = 64 * 1024
 const DEFAULT_MAX_EVENT_CHARACTERS = 64 * 1024
@@ -486,8 +485,8 @@ export class DeepSeekHarnessRuntime implements AgentRuntime {
     )
   }
 
-  private get promptTimeoutMs(): number {
+  private get promptTimeoutMs(): number | undefined {
-    return this.options.promptTimeoutMs ?? DEFAULT_PROMPT_TIMEOUT_MS
+    return this.options.promptTimeoutMs
   }
 
   private get shutdownTimeoutMs(): number {
@@ -1435,6 +1434,8 @@ export class DeepSeekHarnessRuntime implements AgentRuntime {
       } else {
         toolNames.delete(callId)
       }
+      const input = safeStringify(update.rawInput)
+      const output = safeStringify(update.rawOutput)
       return {
         requestId,
         type: 'tool',
@@ -1442,12 +1443,8 @@ export class DeepSeekHarnessRuntime implements AgentRuntime {
         name,
         state,
         summary: `DeepSeek Harness 工具：${name}`,
-        ...(safeStringify(update.rawInput)
+        ...(input ? { input } : {}),
-          ? { input: safeStringify(update.rawInput) }
+        ...(output ? { output } : {})
-          : {}),
-        ...(safeStringify(update.rawOutput)
-          ? { output: safeStringify(update.rawOutput) }
-          : {})
       }
     }
     return undefined
@@ -1505,8 +1502,10 @@ export class DeepSeekHarnessRuntime implements AgentRuntime {
         | { stopReason?: string }
         | undefined
       let promptError: unknown
-      const promptController = new AbortController()
       const cancel = (): void => {
+        run!.closed = true
+        run!.wake?.()
+        run!.wake = undefined
         void state!.agent
           .cancel({
             sessionId: sessionId!
@@ -1514,33 +1513,47 @@ export class DeepSeekHarnessRuntime implements AgentRuntime {
           .catch(() => undefined)
       }
       signal.addEventListener('abort', cancel, { once: true })
-      await withTimeout(
+      if (signal.aborted) {
-        state.agent.extMethod(GOODBUDDY_PREPARE, {
+        cancel()
-          sessionId,
+      }
-          requestId: request.requestId,
+      try {
-          mode:
+        await withTimeout(
-            request.workMode === 'execute' ? 'execute' : 'ask'
+          state.agent.extMethod(GOODBUDDY_PREPARE, {
-        }),
+            sessionId,
-        this.initializationTimeoutMs,
+            requestId: request.requestId,
-        '请求准备'
+            mode:
-      )
+              request.workMode === 'execute' ? 'execute' : 'ask'
-      const prompt = withTimeout(
+          }),
-        state.agent.prompt({
+          this.initializationTimeoutMs,
-          sessionId,
+          '请求准备'
-          prompt: [
+        )
-            {
+        signal.throwIfAborted()
-              type: 'text',
+      } catch (error) {
-              text: flattenPrompt(request)
+        signal.removeEventListener('abort', cancel)
-            },
+        throw error
-            ...(request.images ?? []).map((image) => ({
+      }
-              type: 'image' as const,
+      const promptOperation = state.agent.prompt({
-              data: image.data,
+        sessionId,
-              mimeType: image.mediaType
+        prompt: [
-            }))
+          {
-          ]
+            type: 'text',
-        }),
+            text: flattenPrompt(request)
-        this.promptTimeoutMs,
+          },
-        '请求'
+          ...(request.images ?? []).map((image) => ({
+            type: 'image' as const,
+            data: image.data,
+            mimeType: image.mediaType
+          }))
+        ]
+      })
+      const prompt = (
+        this.promptTimeoutMs === undefined
+          ? promptOperation
+          : withTimeout(
+              promptOperation,
+              this.promptTimeoutMs,
+              '请求'
+            )
       )
         .then((value) => {
           response = value
@@ -1555,9 +1568,15 @@ export class DeepSeekHarnessRuntime implements AgentRuntime {
         })
       try {
         while (!completed || run.updates.length > 0) {
+          signal.throwIfAborted()
           if (this.fatalError) {
             throw this.fatalError
           }
+          if (run.closed && !completed) {
+            throw new Error(
+              'DeepSeek Harness 请求在完成前已关闭'
+            )
+          }
           if (run.updates.length === 0) {
             await this.waitForUpdate(run)
             continue
@@ -1588,11 +1607,21 @@ export class DeepSeekHarnessRuntime implements AgentRuntime {
           sessionId
         }
       } finally {
-        promptController.abort(
+        if (!completed) {
-          new Error('DeepSeek Harness 流式消费已结束')
+          cancel()
-        )
+        }
         signal.removeEventListener('abort', cancel)
-        await prompt
+        if (completed) {
+          await prompt
+        } else {
+          await withTimeout(
+            prompt,
+            this.shutdownTimeoutMs,
+            '取消请求'
+          ).catch(async () => {
+            await this.terminate(state!.child)
+          })
+        }
       }
     } finally {
       signal.removeEventListener('abort', abortTools)

src/main/agent/deepseek-harness-utility-launcher.test.ts

@@ -50,7 +50,7 @@ async function fixture() {
     launchOptions: {
       cwd: workspace,
       signal: new AbortController().signal,
-      baseUrl: 'https://gateway.example/openai/v1',
+      baseUrl: 'http://gateway.example/openai/v1?api-version=1',
       model: 'qwen-plus',
       supportsImageInput: false,
       credentialRefs: [DEEPSEEK_HARNESS_CREDENTIAL_REF],
@@ -102,7 +102,7 @@ describe('DeepSeek Harness utility launcher', () => {
     expect(utility.messages[0]).toMatchObject({
       type: 'start',
       config: {
-        baseUrl: 'https://gateway.example/openai/v1',
+        baseUrl: 'http://gateway.example/openai/v1?api-version=1',
         model: 'qwen-plus',
         supportsImageInput: false,
         credentialRefs: [DEEPSEEK_HARNESS_CREDENTIAL_REF]
@@ -235,10 +235,9 @@ describe('DeepSeek Harness utility launcher', () => {
   })
 
   it.each([
-    'http://gateway.example/v1',
+    'file:///private/config',
-    'https://user:secret@gateway.example/v1',
+    'ftp://gateway.example/v1'
-    'https://gateway.example/v1?api-version=2025-01-01'
+  ])('rejects unsupported endpoint %s before forking', async (baseUrl) => {
-  ])('rejects unsafe endpoint %s before forking', async (baseUrl) => {
     const { dshHome, hostPath, launchOptions } = await fixture()
     const fork = vi.fn()
     const launcher = createDeepSeekHarnessUtilityLauncher({
@@ -250,7 +249,7 @@ describe('DeepSeek Harness utility launcher', () => {
 
     await expect(
       launcher({ ...launchOptions, baseUrl })
-    ).rejects.toThrow('HTTPS')
+    ).rejects.toThrow('HTTP 或 HTTPS')
     expect(fork).not.toHaveBeenCalled()
   })
 })

src/main/agent/deepseek-harness-utility-launcher.ts

@@ -149,7 +149,7 @@ export function createDeepSeekHarnessUtilityLauncher(
     }
     if (!isDeepSeekHarnessCompatibleBaseUrl(options.baseUrl)) {
       throw new Error(
-        'DeepSeek Harness 模型地址必须使用 HTTPS 或本机回环 HTTP，且不得包含凭据、查询参数或片段'
+        'DeepSeek Harness 模型地址必须使用 HTTP 或 HTTPS'
       )
     }
     if (

src/main/agent/opencode-runtime.test.ts

@@ -58,10 +58,6 @@ function fakeClient() {
   } as unknown as ReturnType<typeof createOpencodeClient>
 }
 
-function stdoutOf(child: SpawnedProcess): PassThrough {
-  return child.stdout as PassThrough
-}
-
 function stderrOf(child: SpawnedProcess): PassThrough {
   return child.stderr as PassThrough
 }
@@ -91,6 +87,7 @@ function dependencies(
   spawnMock: ReturnType<typeof vi.fn>
   detectBinary: ReturnType<typeof vi.fn>
   createClient: ReturnType<typeof vi.fn>
+  checkServerHealth: ReturnType<typeof vi.fn>
 } {
   const spawnMock = vi.fn(() => child)
   const detectBinary = vi.fn().mockResolvedValue({
@@ -98,18 +95,21 @@ function dependencies(
     detail: 'OpenCode CLI 已就绪'
   })
   const createClient = vi.fn(() => fakeClient())
+  const checkServerHealth = vi.fn().mockResolvedValue(true)
   return {
     deps: {
       spawn: spawnMock as unknown as typeof spawn,
       detectBinary,
       createClient: createClient as unknown as typeof createOpencodeClient,
+      checkServerHealth,
       platform: 'linux',
       startupTimeoutMs: 100,
       ...overrides
     },
     spawnMock,
     detectBinary,
-    createClient
+    createClient,
+    checkServerHealth
   }
 }
 
@@ -257,19 +257,16 @@ function embeddedRuntime(
   client: ReturnType<typeof createOpencodeClient>,
   overrides: Partial<
     ConstructorParameters<typeof OpenCodeRuntime>[0]
-  > = {}
+  > = {},
+  dependencyOverrides: Partial<OpenCodeRuntimeDependencies> = {}
 ): OpenCodeRuntime {
   const child = fakeChild()
   const { deps } = dependencies(child, {
+    ...dependencyOverrides,
     createClient: vi.fn(
       () => client
     ) as unknown as typeof createOpencodeClient
   })
-  setTimeout(() => {
-    stdoutOf(child).write(
-      'opencode server listening on http://127.0.0.1:4010\n'
-    )
-  }, 0)
   return new OpenCodeRuntime(options(overrides), deps)
 }
 
@@ -301,6 +298,7 @@ describe('OpenCodeRuntime embedded launcher', () => {
       detail: 'OpenCode CLI 已就绪'
     })
     const createClient = vi.fn(() => fakeClient())
+    const checkServerHealth = vi.fn().mockResolvedValue(true)
     const spawnMock = vi.fn((command: string) => {
       if (command === 'taskkill.exe') {
         queueMicrotask(() => {
@@ -308,11 +306,6 @@ describe('OpenCodeRuntime embedded launcher', () => {
         })
         return killerChild
       }
-      setTimeout(() => {
-        stdoutOf(serverChild).write(
-          'opencode server listening securely on http://127.0.0.1:43210\n'
-        )
-      }, 0)
       return serverChild
     })
     const configPath = './private/opencode.json'
@@ -325,6 +318,7 @@ describe('OpenCodeRuntime embedded launcher', () => {
         spawn: spawnMock as unknown as typeof spawn,
         detectBinary,
         createClient: createClient as unknown as typeof createOpencodeClient,
+        checkServerHealth,
         platform: 'win32'
       }
     )
@@ -357,6 +351,14 @@ describe('OpenCodeRuntime embedded launcher', () => {
         })
       })
     )
+    const firstSpawn = spawnMock.mock.calls[0] as unknown as
+      | [string, string[]]
+      | undefined
+    const serverPort = firstSpawn?.[1]
+      ?.find((argument) => argument.startsWith('--port='))
+      ?.slice('--port='.length)
+    expect(serverPort).toMatch(/^\d+$/u)
+    const serverUrl = `http://127.0.0.1:${serverPort}`
     const clientOptions = (
       createClient.mock.calls as unknown as Array<
         [
@@ -375,12 +377,17 @@ describe('OpenCodeRuntime embedded launcher', () => {
         }
       | undefined
     expect(clientOptions).toMatchObject({
-      baseUrl: 'http://127.0.0.1:43210',
+      baseUrl: serverUrl,
       directory: process.cwd(),
       headers: {
         Authorization: expect.stringMatching(/^Basic /u)
       }
     })
+    expect(checkServerHealth).toHaveBeenCalledWith(
+      serverUrl,
+      clientOptions?.headers?.Authorization,
+      expect.any(AbortSignal)
+    )
     const spawnOptions = (
       spawnMock.mock.calls as unknown as Array<
         [string, string[], { env?: NodeJS.ProcessEnv }]
@@ -445,11 +452,6 @@ describe('OpenCodeRuntime embedded launcher', () => {
     )
     const child = fakeChild()
     const { deps, spawnMock } = dependencies(child)
-    setTimeout(() => {
-      stdoutOf(child).write(
-        'opencode server listening on http://127.0.0.1:3012\n'
-      )
-    }, 0)
     const runtime = new OpenCodeRuntime(
       options({
         skillPackages: [
@@ -523,11 +525,6 @@ describe('OpenCodeRuntime embedded launcher', () => {
   it('injects an independent model profile without persisting its key', async () => {
     const child = fakeChild()
     const { deps, spawnMock } = dependencies(child)
-    setTimeout(() => {
-      stdoutOf(child).write(
-        'opencode server listening on http://127.0.0.1:3011\n'
-      )
-    }, 0)
     const runtime = new OpenCodeRuntime(
       options({
         modelProfile: {
@@ -601,11 +598,6 @@ describe('OpenCodeRuntime embedded launcher', () => {
       ])
     )
     Object.assign(process.env, inheritedCredentials)
-    setTimeout(() => {
-      stdoutOf(child).write(
-        'opencode server listening on http://127.0.0.1:3013\n'
-      )
-    }, 0)
     const runtime = new OpenCodeRuntime(
       options({
         modelProfile: {
@@ -730,7 +722,8 @@ describe('OpenCodeRuntime embedded launcher', () => {
             authentication: 'api-key',
             apiKey: 'local-probe-key'
           }
-        })
+        }),
+        { startupTimeoutMs: 20_000 }
       )
       const controller = new AbortController()
       const timeout = setTimeout(
@@ -808,11 +801,6 @@ describe('OpenCodeRuntime embedded launcher', () => {
     }) => {
       const child = fakeChild()
       const { deps, spawnMock } = dependencies(child)
-      setTimeout(() => {
-        stdoutOf(child).write(
-          'opencode server listening on http://127.0.0.1:3012\n'
-        )
-      }, 0)
       const runtime = new OpenCodeRuntime(
         options({
           modelProfile: {
@@ -899,11 +887,6 @@ describe('OpenCodeRuntime embedded launcher', () => {
     process.env.OTEL_EXPORTER_OTLP_ENDPOINT =
       'https://telemetry.invalid'
     try {
-      setTimeout(() => {
-        stdoutOf(child).write(
-          'opencode server listening on http://127.0.0.1:3010\n'
-        )
-      }, 0)
       const runtime = new OpenCodeRuntime(options(), deps)
       await expect(runtime.getStatus()).resolves.toMatchObject({
         available: true
@@ -960,36 +943,39 @@ describe('OpenCodeRuntime embedded launcher', () => {
     }
   })
 
-  it.each([
+  it('polls the known loopback URL without waiting for stdout', async () => {
-    'https://127.0.0.1:4321',
-    'http://0.0.0.0:4321',
-    'http://example.com:4321',
-    'http://127.0.0.1',
-    'http://127.0.0.1:4321/admin'
-  ])('rejects an unsafe listening URL: %s', async (url) => {
     const child = fakeChild()
-    const { deps, createClient } = dependencies(child, {
+    const { deps, checkServerHealth, createClient } = dependencies(child, {
-      spawn: vi.fn(() => {
+      startupTimeoutMs: 1_000
-        queueMicrotask(() => {
-          stdoutOf(child).write(`opencode server listening on ${url}\n`)
-          closeChild(child, 7)
-        })
-        return child
-      }) as unknown as typeof spawn
     })
+    checkServerHealth
+      .mockResolvedValueOnce(false)
+      .mockResolvedValueOnce(true)
     const runtime = new OpenCodeRuntime(options(), deps)
 
     await expect(runtime.getStatus()).resolves.toMatchObject({
-      available: false,
+      available: true
-      detail: 'OpenCode Server 启动前退出（code 7）'
+    })
+    expect(checkServerHealth).toHaveBeenCalledTimes(2)
+    const [url, authorization, signal] =
+      checkServerHealth.mock.calls[0] ?? []
+    expect(url).toMatch(/^http:\/\/127\.0\.0\.1:\d+$/)
+    expect(authorization).toMatch(/^Basic /)
+    expect(signal).toBeInstanceOf(AbortSignal)
+    expect(createClient).toHaveBeenCalledWith({
+      baseUrl: url,
+      directory: process.cwd(),
+      headers: { Authorization: authorization }
     })
-    expect(createClient).not.toHaveBeenCalled()
   })
 
   it('times out, terminates the process tree, and does not expose stderr', async () => {
     const child = fakeChild()
     const secret = 'private-config-token'
-    const { deps } = dependencies(child, { startupTimeoutMs: 5 })
+    const { deps } = dependencies(child, {
+      startupTimeoutMs: 5,
+      checkServerHealth: vi.fn().mockResolvedValue(false)
+    })
     stderrOf(child).write(secret)
     const runtime = new OpenCodeRuntime(options(), deps)
 
@@ -1038,7 +1024,9 @@ describe('OpenCodeRuntime embedded launcher', () => {
 
   it('terminates startup when the request is aborted', async () => {
     const child = fakeChild()
-    const { deps, spawnMock } = dependencies(child)
+    const { deps, spawnMock } = dependencies(child, {
+      checkServerHealth: vi.fn().mockResolvedValue(false)
+    })
     const runtime = new OpenCodeRuntime(options(), deps)
     const controller = new AbortController()
     const stream = runtime.run(
@@ -1058,6 +1046,25 @@ describe('OpenCodeRuntime embedded launcher', () => {
     expect(child.kill).toHaveBeenCalledWith('SIGTERM')
   })
 
+  it('bounds session setup without imposing a full-run deadline', async () => {
+    const harness = runClient([])
+    ;(
+      harness.session.create as unknown as ReturnType<typeof vi.fn>
+    ).mockImplementation(
+      () => new Promise(() => undefined)
+    )
+    const runtime = embeddedRuntime(
+      harness.client,
+      {},
+      { controlRequestTimeoutMs: 5 }
+    )
+
+    await expect(collectRun(runtime)).rejects.toThrow(
+      'OpenCode 创建会话超时'
+    )
+    await runtime.dispose()
+  })
+
   it('keeps external baseUrl mode free of binary detection and spawning', async () => {
     const child = fakeChild()
     const { deps, spawnMock, detectBinary, createClient } = dependencies(child)
@@ -1521,11 +1528,14 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
         ? questionEvent.value.questionId
         : ''
     await runtime.respondToQuestion(questionId, [['先写测试']])
-    expect(setup.questionReply).toHaveBeenCalledWith({
+    expect(setup.questionReply).toHaveBeenCalledWith(
-      requestID: 'question-1',
+      {
-      directory: process.cwd(),
+        requestID: 'question-1',
-      answers: [['先写测试']]
+        directory: process.cwd(),
-    })
+        answers: [['先写测试']]
+      },
+      { signal: expect.any(AbortSignal) }
+    )
     await expect(stream.next()).resolves.toMatchObject({
       value: { type: 'done' }
     })
@@ -1614,20 +1624,212 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
       runtime.respondToQuestion(secondId, [['second answer']])
     ])
     expect(setup.questionReply).toHaveBeenCalledTimes(2)
-    expect(setup.questionReply).toHaveBeenNthCalledWith(1, {
+    expect(setup.questionReply).toHaveBeenNthCalledWith(
-      requestID: 'shared-question',
+      1,
-      directory: process.cwd(),
+      {
-      answers: [['first answer']]
+        requestID: 'shared-question',
-    })
+        directory: process.cwd(),
-    expect(setup.questionReply).toHaveBeenNthCalledWith(2, {
+        answers: [['first answer']]
-      requestID: 'shared-question',
+      },
-      directory: process.cwd(),
+      { signal: expect.any(AbortSignal) }
-      answers: [['second answer']]
+    )
-    })
+    expect(setup.questionReply).toHaveBeenNthCalledWith(
+      2,
+      {
+        requestID: 'shared-question',
+        directory: process.cwd(),
+        answers: [['second answer']]
+      },
+      { signal: expect.any(AbortSignal) }
+    )
     await Promise.all([first.next(), second.next()])
     await runtime.dispose()
   })
 
+  it('emits native Task calls as subagents instead of generic tools', async () => {
+    const taskInput = {
+      subagent_type: 'explorer',
+      description: 'Review application architecture',
+      prompt: 'Inspect the complete source tree.'
+    }
+    const setup = runClient([
+      {
+        id: 'task-running',
+        type: 'message.part.updated',
+        properties: {
+          sessionID: 'session-1',
+          part: {
+            id: 'part-task-1',
+            callID: 'call-task-1',
+            type: 'tool',
+            tool: 'task',
+            state: {
+              status: 'running',
+              input: taskInput,
+              time: { start: 1 }
+            }
+          }
+        }
+      },
+      {
+        id: 'task-completed',
+        type: 'message.part.updated',
+        properties: {
+          sessionID: 'session-1',
+          part: {
+            id: 'part-task-1',
+            callID: 'call-task-1',
+            type: 'tool',
+            tool: 'task',
+            state: {
+              status: 'completed',
+              input: taskInput,
+              output: 'Architecture review complete.',
+              title: 'Review application architecture',
+              metadata: {},
+              time: { start: 1, end: 2 }
+            }
+          }
+        }
+      },
+      {
+        id: 'idle',
+        type: 'session.idle',
+        properties: { sessionID: 'session-1' }
+      }
+    ])
+    const runtime = embeddedRuntime(setup.client)
+
+    const events = await collectRun(runtime)
+    const subagents = events.filter(
+      (event) => event.type === 'subagent'
+    )
+
+    expect(subagents).toHaveLength(2)
+    expect(subagents[0]).toMatchObject({
+      expertName: 'explorer',
+      routingMode: 'native',
+      runtimeCallId: 'call-task-1',
+      state: 'running',
+      reason: 'Review application architecture'
+    })
+    expect(subagents[1]).toMatchObject({
+      childTaskId:
+        subagents[0]?.type === 'subagent'
+          ? subagents[0].childTaskId
+          : undefined,
+      state: 'completed',
+      output: 'Architecture review complete.'
+    })
+    expect(events.some((event) => event.type === 'tool')).toBe(false)
+    await runtime.dispose()
+  })
+
+  it('keeps non-Task tools with Task-shaped input as generic tools', async () => {
+    const setup = runClient([
+      {
+        id: 'custom-tool-completed',
+        type: 'message.part.updated',
+        properties: {
+          sessionID: 'session-1',
+          part: {
+            id: 'part-custom-1',
+            callID: 'call-custom-1',
+            type: 'tool',
+            tool: 'custom_delegate',
+            state: {
+              status: 'completed',
+              input: {
+                subagent_type: 'explorer',
+                prompt: 'Inspect the complete source tree.'
+              },
+              output: 'Custom tool result.',
+              title: 'Custom delegate',
+              metadata: {},
+              time: { start: 1, end: 2 }
+            }
+          }
+        }
+      },
+      {
+        id: 'idle',
+        type: 'session.idle',
+        properties: { sessionID: 'session-1' }
+      }
+    ])
+    const runtime = embeddedRuntime(setup.client)
+
+    const events = await collectRun(runtime)
+
+    expect(events).toContainEqual(
+      expect.objectContaining({
+        type: 'tool',
+        callId: 'call-custom-1',
+        name: 'custom_delegate',
+        state: 'completed'
+      })
+    )
+    expect(events.some((event) => event.type === 'subagent')).toBe(false)
+    await runtime.dispose()
+  })
+
+  it('does not impose a default wall-clock deadline on OpenCode runs', async () => {
+    vi.useFakeTimers()
+    try {
+      let release!: () => void
+      const gate = new Promise<void>((resolve) => {
+        release = resolve
+      })
+      const setup = runClient([])
+      ;(
+        setup.event.subscribe as unknown as ReturnType<typeof vi.fn>
+      ).mockResolvedValue({
+        stream: (async function* () {
+          await gate
+          yield {
+            id: 'idle',
+            type: 'session.idle',
+            properties: { sessionID: 'session-1' }
+          }
+        })()
+      })
+      const runtime = new OpenCodeRuntime(
+        options({
+          baseUrl: 'http://127.0.0.1:4096',
+          embedded: false
+        }),
+        dependencies(fakeChild(), {
+          createClient: vi.fn(
+            () => setup.client
+          ) as unknown as typeof createOpencodeClient
+        }).deps
+      )
+      const stream = runtime.run(
+        {
+          requestId: '3f496642-f47d-4e0a-8944-a32c77b0d6ef',
+          conversationId: 'conversation-1',
+          prompt: 'work for more than a day',
+          workMode: 'execute'
+        },
+        new AbortController().signal
+      )
+
+      await expect(stream.next()).resolves.toMatchObject({
+        value: { type: 'status' }
+      })
+      const pending = stream.next()
+      await vi.advanceTimersByTimeAsync(25 * 60 * 60_000)
+      expect(setup.session.abort).not.toHaveBeenCalled()
+      release()
+      await expect(pending).resolves.toMatchObject({
+        value: { type: 'done' }
+      })
+      await runtime.dispose()
+    } finally {
+      vi.useRealTimers()
+    }
+  })
+
   it('aborts an OpenCode run at its total execution deadline', async () => {
     const setup = runClient([])
     ;(
@@ -1705,11 +1907,6 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
         executionTimeoutMs: 20
       }).deps
     )
-    setTimeout(() => {
-      stdoutOf(child).write(
-        'opencode server listening on http://127.0.0.1:4010\n'
-      )
-    }, 0)
     const stream = runtime.run(
       {
         requestId: '3f496642-f47d-4e0a-8944-a32c77b0d6ef',
@@ -1886,11 +2083,6 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
         () => setup.client
       ) as unknown as typeof createOpencodeClient
     })
-    setTimeout(() => {
-      stdoutOf(child).write(
-        'opencode server listening on http://127.0.0.1:4010\n'
-      )
-    }, 0)
     const runtime = new OpenCodeRuntime(
       options({ knowledgeGateway: gateway }),
       deps
@@ -1987,11 +2179,6 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
         () => setup.client
       ) as unknown as typeof createOpencodeClient
     })
-    setTimeout(() => {
-      stdoutOf(child).write(
-        'opencode server listening on http://127.0.0.1:4010\n'
-      )
-    }, 0)
     const runtime = new OpenCodeRuntime(
       options({
         knowledgeGateway: {
@@ -2090,11 +2277,6 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
         () => setup.client
       ) as unknown as typeof createOpencodeClient
     })
-    setTimeout(() => {
-      stdoutOf(child).write(
-        'opencode server listening on http://127.0.0.1:4010\n'
-      )
-    }, 0)
     const runtime = new OpenCodeRuntime(
       options({
         knowledgeGateway: {
@@ -2340,11 +2522,14 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
       { signal: expect.any(AbortSignal) }
     )
     expect(permissionReply).toHaveBeenCalledOnce()
-    expect(permissionReply).toHaveBeenCalledWith({
+    expect(permissionReply).toHaveBeenCalledWith(
-      requestID: 'permission-1',
+      {
-      directory: process.cwd(),
+        requestID: 'permission-1',
-      reply: 'once'
+        directory: process.cwd(),
-    })
+        reply: 'once'
+      },
+      { signal: expect.any(AbortSignal) }
+    )
     expect(events).toContainEqual(
       expect.objectContaining({
         type: 'reasoning',
@@ -2436,14 +2621,16 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
           requestID: 'permission-1',
           directory: process.cwd(),
           reply: 'once'
-        }
+        },
+        { signal: expect.any(AbortSignal) }
       ],
       [
         {
           requestID: 'permission-2',
           directory: process.cwd(),
           reply: 'once'
-        }
+        },
+        { signal: expect.any(AbortSignal) }
       ]
     ])
     await runtime.dispose()
@@ -2609,10 +2796,13 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
     await collectRun(runtime)
     await runtime.releaseConversation('conversation-1')
 
-    expect(session.delete).toHaveBeenCalledWith({
+    expect(session.delete).toHaveBeenCalledWith(
-      sessionID: 'session-1',
+      {
-      directory: process.cwd()
+        sessionID: 'session-1',
-    })
+        directory: process.cwd()
+      },
+      { signal: expect.any(AbortSignal) }
+    )
     await runtime.dispose()
   })
 
@@ -2632,11 +2822,14 @@ describe('OpenCodeRuntime embedded permission mediation', () => {
     await collectRun(runtime, 'execute')
 
     expect(permissionReply).toHaveBeenCalledOnce()
-    expect(permissionReply).toHaveBeenCalledWith({
+    expect(permissionReply).toHaveBeenCalledWith(
-      requestID: 'permission-1',
+      {
-      directory: process.cwd(),
+        requestID: 'permission-1',
-      reply: 'reject'
+        directory: process.cwd(),
-    })
+        reply: 'reject'
+      },
+      { signal: expect.any(AbortSignal) }
+    )
     await runtime.dispose()
   })
 

src/main/agent/opencode-subagent.test.ts

@@ -0,0 +1,57 @@
+import { describe, expect, it } from 'vitest'
+import { toOpenCodeSubagentEvent } from './opencode-subagent'
+
+const requestId = '3f496642-f47d-4e0a-8944-a32c77b0d6ef'
+
+describe('OpenCode native subagent events', () => {
+  it('maps Task tool updates to stable bounded subagent identities', () => {
+    const running = toOpenCodeSubagentEvent({
+      requestId,
+      callId: 'call-task-1',
+      state: 'running',
+      input: {
+        subagent_type: 'explorer',
+        description: 'Review application architecture',
+        prompt: 'Inspect the complete source tree.'
+      }
+    })
+    const completed = toOpenCodeSubagentEvent({
+      requestId,
+      callId: 'call-task-1',
+      state: 'completed',
+      input: {
+        subagent_type: 'explorer',
+        description: 'Review application architecture',
+        prompt: 'Inspect the complete source tree.'
+      },
+      output: 'Architecture review complete.'
+    })
+
+    expect(running).toMatchObject({
+      type: 'subagent',
+      expertName: 'explorer',
+      routingMode: 'native',
+      runtimeCallId: 'call-task-1',
+      state: 'running',
+      reason: 'Review application architecture'
+    })
+    expect(completed).toMatchObject({
+      childTaskId: running?.childTaskId,
+      expertId: running?.expertId,
+      state: 'completed',
+      output: 'Architecture review complete.'
+    })
+  })
+
+  it('does not classify unrelated tools as subagents', () => {
+    expect(
+      toOpenCodeSubagentEvent({
+        requestId,
+        callId: 'call-read-1',
+        state: 'completed',
+        input: { filePath: 'README.md' },
+        output: 'contents'
+      })
+    ).toBeUndefined()
+  })
+})

src/main/agent/opencode-subagent.ts

@@ -0,0 +1,140 @@
+import { createHash } from 'node:crypto'
+import type { SubagentEvent } from '../../shared/contracts'
+import { boundedToolDetail } from './approval-summary'
+
+type OpenCodeSubagentState =
+  | 'pending'
+  | 'running'
+  | 'in_progress'
+  | 'completed'
+  | 'failed'
+  | 'error'
+
+export type OpenCodeSubagentToolUpdate = {
+  requestId: string
+  callId: string
+  state: OpenCodeSubagentState
+  input: unknown
+  output?: unknown
+  error?: unknown
+}
+
+export type OpenCodeSubagentInput = {
+  subagent_type: string
+  prompt: string
+  description?: string
+  command?: string
+}
+
+function isRecord(value: unknown): value is Record<string, unknown> {
+  return (
+    typeof value === 'object' &&
+    value !== null &&
+    !Array.isArray(value)
+  )
+}
+
+function boundedText(
+  value: unknown,
+  maximum: number
+): string | undefined {
+  if (typeof value !== 'string') {
+    return undefined
+  }
+  const normalized = value.trim()
+  return normalized ? normalized.slice(0, maximum) : undefined
+}
+
+function boundedDetail(
+  value: unknown,
+  maximum: number
+): string | undefined {
+  if (typeof value === 'string') {
+    return boundedText(value, maximum)
+  }
+  if (value === undefined || value === null) {
+    return undefined
+  }
+  return boundedToolDetail(value, maximum)
+}
+
+function deterministicUuid(value: string): string {
+  const bytes = createHash('sha256').update(value).digest().subarray(0, 16)
+  bytes[6] = (bytes[6]! & 0x0f) | 0x50
+  bytes[8] = (bytes[8]! & 0x3f) | 0x80
+  const hex = bytes.toString('hex')
+  return [
+    hex.slice(0, 8),
+    hex.slice(8, 12),
+    hex.slice(12, 16),
+    hex.slice(16, 20),
+    hex.slice(20)
+  ].join('-')
+}
+
+export function parseOpenCodeSubagentInput(
+  value: unknown
+): OpenCodeSubagentInput | undefined {
+  if (!isRecord(value)) {
+    return undefined
+  }
+  const subagentType = boundedText(value.subagent_type, 80)
+  const prompt = boundedText(value.prompt, 20_000)
+  if (!subagentType || !prompt) {
+    return undefined
+  }
+  const description = boundedText(value.description, 240)
+  const command = boundedText(value.command, 240)
+  return {
+    subagent_type: subagentType,
+    prompt,
+    ...(description ? { description } : {}),
+    ...(command ? { command } : {})
+  }
+}
+
+export function toOpenCodeSubagentEvent(
+  update: OpenCodeSubagentToolUpdate
+): SubagentEvent | undefined {
+  const input = parseOpenCodeSubagentInput(update.input)
+  if (!input) {
+    return undefined
+  }
+  const expertName = input.subagent_type
+  const description =
+    input.description ?? input.command
+  const state =
+    update.state === 'pending'
+      ? 'queued'
+      : update.state === 'running' ||
+          update.state === 'in_progress'
+        ? 'running'
+        : update.state === 'completed'
+          ? 'completed'
+          : 'failed'
+  const failed = state === 'failed'
+  const output = failed
+    ? undefined
+    : boundedDetail(update.output, 16_000)
+  const error = failed
+    ? boundedDetail(update.error ?? update.output, 1_000)
+    : undefined
+
+  return {
+    requestId: update.requestId,
+    type: 'subagent',
+    childTaskId: deterministicUuid(
+      `opencode-subagent-call\0${update.requestId}\0${update.callId}`
+    ),
+    expertId: deterministicUuid(
+      `opencode-subagent-type\0${expertName}`
+    ),
+    expertName,
+    routingMode: 'native',
+    runtimeCallId: update.callId.slice(0, 256),
+    state,
+    ...(description ? { reason: description } : {}),
+    ...(output ? { output } : {}),
+    ...(error ? { error } : {})
+  }
+}

src/main/agent/runtime-selection.ts

@@ -139,7 +139,7 @@ export function applyRuntimeSelection(
       !isDeepSeekHarnessModelProfile(selectedProfile)
     ) {
       throw new Error(
-        'DeepSeek Harness 仅支持使用 API Key 的安全 OpenAI 兼容 Chat Completions 连接'
+        'DeepSeek Harness 仅支持使用 API Key 的 OpenAI 兼容 Chat Completions 连接'
       )
     }
     return {

src/main/assistant/assistant-database.test.ts

@@ -3308,6 +3308,20 @@ describe('AssistantDatabase', () => {
                 output: '远程输出'.repeat(10_000)
               }
             ],
+            subagents: [
+              {
+                childTaskId:
+                  '00000000-0000-4000-8000-000000000513',
+                expertId:
+                  '00000000-0000-4000-8000-000000000514',
+                expertName: 'explorer',
+                routingMode: 'native' as const,
+                runtimeCallId: 'call-task-1',
+                state: 'completed' as const,
+                reason: 'Review application architecture',
+                output: 'Architecture review complete.'
+              }
+            ],
             contextCompressions: [
               {
                 state: 'completed' as const,
@@ -3368,6 +3382,20 @@ describe('AssistantDatabase', () => {
               output: '远程输出'.repeat(10_000)
             }
           ],
+          subagents: [
+            {
+              childTaskId:
+                '00000000-0000-4000-8000-000000000513',
+              expertId:
+                '00000000-0000-4000-8000-000000000514',
+              expertName: 'explorer',
+              routingMode: 'native',
+              runtimeCallId: 'call-task-1',
+              state: 'completed',
+              reason: 'Review application architecture',
+              output: 'Architecture review complete.'
+            }
+          ],
           contextCompressions: [
             {
               state: 'completed',
@@ -4278,7 +4306,7 @@ describe('AssistantDatabase', () => {
     database.close()
   })
 
-  it('durably interrupts active tool metadata during startup recovery', async () => {
+  it('durably interrupts active tool and subagent metadata during startup recovery', async () => {
     const directory = await mkdtemp(
       join(tmpdir(), 'goodbuddy-conversation-recovery-')
     )
@@ -4289,6 +4317,8 @@ describe('AssistantDatabase', () => {
     const messageId = '00000000-0000-4000-8000-000000000218'
     const cancelledMessageId =
       '00000000-0000-4000-8000-000000000219'
+    const subagentMessageId =
+      '00000000-0000-4000-8000-000000000220'
     const initial = new AssistantDatabase(databasePath)
     initial.initialize('C:\\Workspace')
     initial.replaceConversations([
@@ -4342,6 +4372,38 @@ describe('AssistantDatabase', () => {
                 error: 'runtime parser detail'
               }
             ]
+          },
+          {
+            id: subagentMessageId,
+            role: 'assistant',
+            content: '父请求已完成',
+            createdAt: 1_775_000_003_000,
+            state: 'complete',
+            status: '已完成',
+            subagents: [
+              {
+                childTaskId:
+                  '00000000-0000-4000-8000-000000000221',
+                expertId:
+                  '00000000-0000-4000-8000-000000000222',
+                expertName: 'explorer',
+                routingMode: 'native',
+                runtimeCallId: 'task-1',
+                state: 'running',
+                reason: '检查项目'
+              },
+              {
+                childTaskId:
+                  '00000000-0000-4000-8000-000000000223',
+                expertId:
+                  '00000000-0000-4000-8000-000000000224',
+                expertName: 'worker',
+                routingMode: 'native',
+                runtimeCallId: 'task-2',
+                state: 'completed',
+                output: '已完成'
+              }
+            ]
           }
         ]
       }
@@ -4370,6 +4432,21 @@ describe('AssistantDatabase', () => {
         expect.objectContaining({ name: 'failed-tool', state: 'failed' })
       ]
     })
+    expect(recovered.listConversations()[0]?.messages[2]).toMatchObject({
+      id: subagentMessageId,
+      state: 'complete',
+      status: '已完成',
+      subagents: [
+        expect.objectContaining({
+          runtimeCallId: 'task-1',
+          state: 'cancelled'
+        }),
+        expect.objectContaining({
+          runtimeCallId: 'task-2',
+          state: 'completed'
+        })
+      ]
+    })
     expect(recovered.listConversations()[0]?.messages[1]).toMatchObject({
       id: cancelledMessageId,
       state: 'error',

src/main/assistant/assistant-database.ts

@@ -215,6 +215,7 @@ type MessageMetadata = {
   contextCompression?: ConversationSnapshot['messages'][number]['contextCompression']
   contextCompressions?: ConversationSnapshot['messages'][number]['contextCompressions']
   tools?: ConversationSnapshot['messages'][number]['tools']
+  subagents?: ConversationSnapshot['messages'][number]['subagents']
   sources?: string[]
   sourceReferences?: ConversationSnapshot['messages'][number]['sourceReferences']
   knowledgeRetrieval?: ConversationSnapshot['messages'][number]['knowledgeRetrieval']
@@ -903,6 +904,16 @@ function interruptActiveTools(
   )
 }
 
+function interruptActiveSubagents(
+  subagents: MessageMetadata['subagents']
+): MessageMetadata['subagents'] {
+  return subagents?.map((subagent) =>
+    subagent.state === 'queued' || subagent.state === 'running'
+      ? { ...subagent, state: 'cancelled' as const }
+      : subagent
+  )
+}
+
 function interruptActiveToolBlocks(
   blocks: MessageMetadata['blocks']
 ): MessageMetadata['blocks'] {
@@ -1017,6 +1028,9 @@ function toConversationSnapshot(
         tools: interrupted
           ? interruptActiveTools(metadata.tools)
           : metadata.tools,
+        subagents: interrupted
+          ? interruptActiveSubagents(metadata.subagents)
+          : metadata.subagents,
         sources: metadata.sources,
         sourceReferences: metadata.sourceReferences,
         knowledgeRetrieval: metadata.knowledgeRetrieval,
@@ -1040,6 +1054,7 @@ function serializeConversationMessageMetadata(
     contextCompression: message.contextCompression,
     contextCompressions: message.contextCompressions,
     tools: message.tools,
+    subagents: message.subagents,
     sources: message.sources,
     sourceReferences: message.sourceReferences,
     knowledgeRetrieval: message.knowledgeRetrieval,
@@ -1417,7 +1432,7 @@ export class AssistantDatabase {
           const metadata = JSON.parse(
             message.metadata_json
           ) as MessageMetadata
-          const hasActiveTool = Boolean(
+          const hasActiveActivity = Boolean(
             metadata.tools?.some(
               (tool) =>
                 tool.state === 'pending' || tool.state === 'running'
@@ -1427,9 +1442,17 @@ export class AssistantDatabase {
                   block.type === 'tool' &&
                   (block.tool.state === 'pending' ||
                     block.tool.state === 'running')
+              ) ||
+              metadata.subagents?.some(
+                (subagent) =>
+                  subagent.state === 'queued' ||
+                  subagent.state === 'running'
               )
           )
-          if (message.state !== 'streaming' && !hasActiveTool) {
+          if (
+            message.state !== 'streaming' &&
+            !hasActiveActivity
+          ) {
             continue
           }
           updateMessage.run(
@@ -1441,7 +1464,10 @@ export class AssistantDatabase {
                   ? interruptedMessageStatus
                   : metadata.status,
               tools: interruptActiveTools(metadata.tools),
-              blocks: interruptActiveToolBlocks(metadata.blocks)
+              blocks: interruptActiveToolBlocks(metadata.blocks),
+              subagents: interruptActiveSubagents(
+                metadata.subagents
+              )
             }),
             message.id
           )

src/main/channels/channel-settings-store.test.ts

@@ -231,7 +231,7 @@ describe('ChannelSettingsStore', () => {
     const bound = await store.saveWeixinBinding({
       accountId: 'account-123456',
       userId: 'user-654321',
-      baseUrl: 'https://ilinkai.weixin.qq.com',
+      baseUrl: 'http://ilinkai.weixin.qq.com',
       token: 'weixin-private-token'
     })
     expect(bound.weixin).toMatchObject({

src/main/channels/channel-settings-store.ts

@@ -300,9 +300,13 @@ const weixinBindingSchema = z
       .string()
       .url()
       .max(2_048)
-      .refine((value) => new URL(value).protocol === 'https:', {
+      .refine(
-        message: '微信服务地址必须使用 HTTPS'
+        (value) =>
-      }),
+          ['http:', 'https:'].includes(new URL(value).protocol),
+        {
+          message: '微信服务地址必须使用 HTTP 或 HTTPS'
+        }
+      ),
     token: z
       .string()
       .trim()

src/main/deepseek-harness-host.test.ts

@@ -39,7 +39,7 @@ describe('controlled DeepSeek Harness host', () => {
         dshHome: 'C:\\controlled-dsh-home',
         skillPackages: []
       })
-    ).rejects.toThrow('secure OpenAI-compatible')
+    ).rejects.toThrow('HTTP or HTTPS OpenAI-compatible')
   })
 
   it('suppresses console payloads instead of contaminating stdout', () => {

src/main/deepseek-harness-host.ts

@@ -93,7 +93,7 @@ function validateHostConfig(
     !isDeepSeekHarnessCompatibleBaseUrl(config.baseUrl)
   ) {
     throw new Error(
-      'Controlled Harness requires a secure OpenAI-compatible Chat Completions endpoint'
+      'Controlled Harness requires an HTTP or HTTPS OpenAI-compatible Chat Completions endpoint'
     )
   }
   if (!config.credentialRefs.length) {

src/main/ipc.test.ts

@@ -6706,6 +6706,62 @@ describe('registerIpcHandlers agent terminal state', () => {
     await harness.dispose()
   })
 
+  it('does not fail a completed native subagent for its provisional Task tool', async () => {
+    const runtime = {
+      capability: 'chat',
+      requiresToolApproval: false,
+      supportsToolExecution: true,
+      getStatus: vi.fn(),
+      dispose: vi.fn(),
+      async *run(request: { requestId: string }) {
+        yield {
+          requestId: request.requestId,
+          type: 'tool',
+          callId: 'call-task-1',
+          name: 'task',
+          state: 'running',
+          summary: 'OpenCode 工具：task'
+        }
+        yield {
+          requestId: request.requestId,
+          type: 'subagent',
+          childTaskId: '00000000-0000-4000-8000-000000000601',
+          expertId: '00000000-0000-4000-8000-000000000701',
+          expertName: 'explore',
+          routingMode: 'native',
+          runtimeCallId: 'call-task-1',
+          state: 'completed',
+          output: 'Desktop files inspected.'
+        }
+        yield { requestId: request.requestId, type: 'done' }
+      }
+    }
+    const harness = createHarness(runtime)
+    const requestId = '3f496642-f47d-4e0a-8944-a32c77b0d6ef'
+
+    harness.handler?.(trustedEvent(harness.webContents), {
+      requestId,
+      conversationId: 'conversation-1',
+      prompt: 'inspect the desktop',
+      workMode: 'execute'
+    })
+
+    await vi.waitFor(() =>
+      expect(harness.assistantDatabase.updateTaskStatus).toHaveBeenCalledWith(
+        requestId,
+        'completed'
+      )
+    )
+    expect(
+      harness.assistantDatabase.updateTaskStatus
+    ).not.toHaveBeenCalledWith(
+      requestId,
+      'failed',
+      expect.any(String)
+    )
+    await harness.dispose()
+  })
+
   it.each(['opencode', 'continue'] as const)(
     'preserves read-only Ask mode at the %s Runtime boundary',
     async (runtimeId) => {

src/main/ipc.ts

@@ -3488,6 +3488,13 @@ export function registerIpcHandlers(
           if (publicEvent.type === 'tool') {
             toolStates.set(publicEvent.callId, publicEvent)
           }
+          if (
+            publicEvent.type === 'subagent' &&
+            publicEvent.routingMode === 'native' &&
+            publicEvent.runtimeCallId
+          ) {
+            toolStates.delete(publicEvent.runtimeCallId)
+          }
           if (publicEvent.type === 'question') {
             pendingAgentQuestions.set(publicEvent.questionId, {
               requestId: request.requestId,

src/main/release-notes-service.test.ts

@@ -88,6 +88,16 @@ describe('ReleaseNotesService', () => {
     })
   })
 
+  it('does not present older notes as a missing current release', async () => {
+    const { service, settingsStore } = await createService('0.8.19')
+    await settingsStore.setLastSeenReleaseNotesVersion('0.8.11')
+
+    await expect(service.getPending()).resolves.toEqual({
+      currentVersion: '0.8.19',
+      releases: []
+    })
+  })
+
   it('persists acknowledgement and does not show the release again', async () => {
     const { service, settingsStore } = await createService('0.8.18')
 

src/main/release-notes-service.ts

@@ -57,6 +57,12 @@ export class ReleaseNotesService {
   async getPending(): Promise<ReleaseNotesSnapshot> {
     const releases = await this.loadReleases()
     const currentVersion = this.dependencies.currentVersion
+    if (!releases.some((release) => release.version === currentVersion)) {
+      return {
+        currentVersion,
+        releases: []
+      }
+    }
     const lastSeenVersion =
       await this.dependencies.settingsStore.getLastSeenReleaseNotesVersion()
     const pending = releases.filter((release) => {

src/main/release-notes-source.test.ts

@@ -11,8 +11,15 @@ describe('packaged release notes', () => {
         'utf8'
       )
     ) as unknown
+    const packageSource = JSON.parse(
+      await readFile(join(process.cwd(), 'package.json'), 'utf8')
+    ) as { version?: unknown }
     const parsed = releaseNotesFileSchema.parse(source)
 
+    expect(typeof packageSource.version).toBe('string')
+    expect(parsed.releases).toContainEqual(
+      expect.objectContaining({ version: packageSource.version })
+    )
     expect(parsed.releases).toContainEqual(
       expect.objectContaining({ version: '0.8.19' })
     )

src/main/remote-agent/protocol-remote-runtime-channel.ts

@@ -22,7 +22,8 @@ import { AgentFrameError } from '../../shared/agent-protocol/frame'
 import {
   assertRemotePromptAcceptanceMatchesPreparation,
   remotePromptOperationAcceptanceSchema,
-  remotePromptOperationPreparationSchema
+  remotePromptOperationPreparationSchema,
+  UNBOUNDED_REMOTE_PROMPT_DEADLINE
 } from '../../shared/remote-agent-contracts'
 import {
   modelBridgePolicySchema
@@ -66,6 +67,9 @@ const RUNTIME_MODEL_BRIDGE_CAPABILITY = 'runtime/model-bridge'
 const RUNTIME_MODEL_BRIDGE_CAPABILITY_VERSION = 1
 const DEFAULT_CONTROL_TIMEOUT_MS = 15_000
 const DEFAULT_MAXIMUM_PENDING_CONTROL_REQUESTS = 32
+const UNBOUNDED_REMOTE_PROMPT_DEADLINE_MS = Date.parse(
+  UNBOUNDED_REMOTE_PROMPT_DEADLINE
+)
 const MAXIMUM_PENDING_OUTPUT_WRITES = 256
 const MAXIMUM_OUTPUT_WRITE_BYTES =
   AGENT_PROTOCOL_LIMITS.maximumBufferedProtocolBytes
@@ -1038,11 +1042,16 @@ export class ProtocolRemoteRuntimeChannel
     ].some(
       (bridge) => bridge.providerDeliveryMayBeUncertain
     )
+    const recoveryDeadlineExpired =
+      this.#recoveryDeadlineAt !== undefined &&
+      this.#recoveryDeadlineAt !==
+        UNBOUNDED_REMOTE_PROMPT_DEADLINE_MS &&
+      Date.now() >= this.#recoveryDeadlineAt
     if (
       this.#recoveryDeadlineAt === undefined ||
       this.#recoverySignal === undefined ||
       this.#recoverySignal.aborted ||
-      Date.now() >= this.#recoveryDeadlineAt ||
+      recoveryDeadlineExpired ||
       this.#activeBinarySends > 0 ||
       this.#pendingControlRequests > 0 ||
       providerDeliveryMayBeUncertain ||
@@ -1090,11 +1099,13 @@ export class ProtocolRemoteRuntimeChannel
     await this.#flushRecoveryPersistence()
     const deadlineAt = this.#recoveryDeadlineAt!
     const callerSignal = this.#recoverySignal!
-    const remaining = deadlineAt - Date.now()
+    if (
-    if (remaining <= 0) {
+      deadlineAt !== UNBOUNDED_REMOTE_PROMPT_DEADLINE_MS &&
+      Date.now() >= deadlineAt
+    ) {
       throw new ProtocolRemoteRuntimeChannelError(
         'Remote Runtime recovery deadline expired',
-        'aborted'
+        'transport'
       )
     }
     const recoveryController = new AbortController()
@@ -1108,6 +1119,13 @@ export class ProtocolRemoteRuntimeChannel
       )
     }
     callerSignal.addEventListener('abort', abort, { once: true })
+    const recoveryTimeoutMs =
+      deadlineAt === UNBOUNDED_REMOTE_PROMPT_DEADLINE_MS
+        ? this.#state.controlTimeoutMs
+        : Math.min(
+            this.#state.controlTimeoutMs,
+            Math.max(0, deadlineAt - Date.now())
+          )
     const timeout = setTimeout(() => {
       recoveryController.abort(
         new DOMException(
@@ -1115,7 +1133,7 @@ export class ProtocolRemoteRuntimeChannel
           'TimeoutError'
         )
       )
-    }, remaining)
+    }, recoveryTimeoutMs)
     try {
       await this.#state.connection.reconnect!(
         recoveryController.signal
@@ -1151,10 +1169,7 @@ export class ProtocolRemoteRuntimeChannel
             channelId: this.#state.openResult.channelId,
             channelEpoch: this.#state.openResult.channelEpoch
           },
-          Math.min(
+          this.#state.controlTimeoutMs,
-            this.#state.controlTimeoutMs,
-            Math.max(1, deadlineAt - Date.now())
-          ),
           recoveryController.signal
         ),
         'Remote Runtime resume response is invalid'
@@ -1205,10 +1220,7 @@ export class ProtocolRemoteRuntimeChannel
             acknowledgedSequence:
               resume.cursors.lastMainAckSequence
           },
-          Math.min(
+          this.#state.controlTimeoutMs,
-            this.#state.controlTimeoutMs,
-            Math.max(1, deadlineAt - Date.now())
-          ),
           recoveryController.signal
         ),
         'Remote Runtime replay response is invalid'

src/main/runtime-settings-store.test.ts

@@ -458,7 +458,7 @@ describe('RuntimeSettingsStore', () => {
     const { store } = await createStore({
       GOODBUDDY_MODEL_API_KEY: apiKey,
       GOODBUDDY_MODEL_BASE_URL:
-        'https://gateway.example/openai/v1',
+        'http://gateway.example/openai/v1?api-version=1',
       GOODBUDDY_MODEL_NAME: 'qwen-plus'
     })
 
@@ -467,7 +467,7 @@ describe('RuntimeSettingsStore', () => {
       deepseekHarnessModelProfile: {
         id: 'goodbuddy-platform-harness',
         name: '管理员预置模型',
-        baseUrl: 'https://gateway.example/openai/v1',
+        baseUrl: 'http://gateway.example/openai/v1?api-version=1',
         modelName: 'qwen-plus',
         protocol: 'openai-chat-completions',
         authentication: 'api-key',
@@ -486,28 +486,11 @@ describe('RuntimeSettingsStore', () => {
 
   it.each([
     [
-      'an insecure public endpoint',
+      'an unsupported endpoint protocol',
-      {
-        GOODBUDDY_MODEL_API_KEY: 'platform-key',
-        GOODBUDDY_MODEL_BASE_URL: 'http://gateway.example/v1',
-        GOODBUDDY_MODEL_NAME: 'qwen-plus'
-      }
-    ],
-    [
-      'an endpoint with embedded credentials',
       {
         GOODBUDDY_MODEL_API_KEY: 'platform-key',
         GOODBUDDY_MODEL_BASE_URL:
-          'https://user:secret@gateway.example/v1',
+          'ftp://gateway.example/v1',
-        GOODBUDDY_MODEL_NAME: 'qwen-plus'
-      }
-    ],
-    [
-      'an endpoint with a query string',
-      {
-        GOODBUDDY_MODEL_API_KEY: 'platform-key',
-        GOODBUDDY_MODEL_BASE_URL:
-          'https://gateway.example/v1?api-version=1',
         GOODBUDDY_MODEL_NAME: 'qwen-plus'
       }
     ],

src/main/runtime-settings-store.ts

@@ -2290,7 +2290,7 @@ export class RuntimeSettingsStore {
       }
       if (!isDeepSeekHarnessModelProfile(profile)) {
         throw new Error(
-          'DeepSeek Harness 仅支持使用 API Key 的安全 OpenAI 兼容 Chat Completions 连接'
+          'DeepSeek Harness 仅支持使用 API Key 的 OpenAI 兼容 Chat Completions 连接'
         )
       }
     }

src/renderer/src/App.test.tsx

@@ -8515,7 +8515,7 @@ describe('App', () => {
     )
   })
 
-  it('shows bounded Subagent states and records child expert activity', async () => {
+  it('shows Subagent states and records child expert activity', async () => {
     render(<App />)
     fireEvent.change(await screen.findByLabelText('向 GoodBuddy 提问'), {
       target: { value: '分析复杂问题' }
@@ -8568,19 +8568,26 @@ describe('App', () => {
       }
     })
 
-    const statusRegion = await screen.findByLabelText('子专家状态')
+    const statusRegion = await screen.findByLabelText('子代理状态')
     expect(within(statusRegion).getByText('研究专家')).toBeInTheDocument()
     expect(within(statusRegion).getByText('等待中')).toBeInTheDocument()
     expect(within(statusRegion).getByText('代码专家')).toBeInTheDocument()
     expect(within(statusRegion).getByText('进行中')).toBeInTheDocument()
     expect(within(statusRegion).getByText('安全专家')).toBeInTheDocument()
     expect(within(statusRegion).getByText('失败')).toBeInTheDocument()
+    const subagentCards = within(statusRegion).getAllByRole('group')
+    fireEvent.click(
+      within(subagentCards[2]!).getByText('安全专家', {
+        selector: 'strong'
+      })
+    )
+    fireEvent(subagentCards[2]!, new Event('toggle'))
     expect(
       within(statusRegion).getByText('无法读取必要上下文')
     ).toBeInTheDocument()
     expect(
-      within(statusRegion).queryByText('第四位专家')
+      within(statusRegion).getByText('第四位专家')
-    ).not.toBeInTheDocument()
+    ).toBeInTheDocument()
 
     act(() => {
       agentListener?.({
@@ -8599,9 +8606,20 @@ describe('App', () => {
         output: '代码专家的部分结论'
       })
     })
-    expect(within(statusRegion).getByText('已完成')).toBeInTheDocument()
+    expect(within(statusRegion).getAllByText('已完成')).toHaveLength(2)
     expect(within(statusRegion).getByText('已取消')).toBeInTheDocument()
     expect(within(statusRegion).getByText('父任务已停止')).toBeInTheDocument()
+    for (const [index, expertName] of [
+      '研究专家',
+      '代码专家'
+    ].entries()) {
+      fireEvent.click(
+        within(subagentCards[index]!).getByText(expertName, {
+          selector: 'strong'
+        })
+      )
+      fireEvent(subagentCards[index]!, new Event('toggle'))
+    }
     expect(
       within(statusRegion).getByText('研究专家的独立结论')
     ).toBeInTheDocument()
@@ -8627,7 +8645,7 @@ describe('App', () => {
     fireEvent.click(screen.getByText('运行记录'))
     expect(await screen.findAllByText('子专家')).toHaveLength(4)
     expect(screen.getAllByText(/智能路由/u).length).toBeGreaterThan(0)
-    expect(screen.getAllByText(/手动指定/u).length).toBeGreaterThan(0)
+    expect(screen.getAllByText(/指定专家/u).length).toBeGreaterThan(0)
     expect(
       screen.getByText(/研究专家的独立结论/u)
     ).toBeInTheDocument()
@@ -8636,6 +8654,144 @@ describe('App', () => {
     ).toBeInTheDocument()
   })
 
+  it('replaces an incremental OpenCode Task tool with a native subagent card', async () => {
+    render(<App />)
+    fireEvent.change(await screen.findByLabelText('向 GoodBuddy 提问'), {
+      target: { value: '检查架构' }
+    })
+    fireEvent.click(screen.getByLabelText('发送'))
+    await waitFor(() => expect(run).toHaveBeenCalledOnce())
+    const request = run.mock.calls[0]?.[0]
+    if (!request) {
+      throw new Error('Missing request')
+    }
+
+    act(() => {
+      agentListener?.({
+        requestId: request.requestId,
+        type: 'tool',
+        callId: 'call-task-1',
+        name: 'Review application architecture',
+        state: 'running',
+        summary: '远端 Runtime 工具：Review application architecture'
+      })
+      agentListener?.({
+        requestId: request.requestId,
+        type: 'reasoning',
+        delta: '等待子代理结果'
+      })
+    })
+    expect(await screen.findByText('等待子代理结果')).toBeInTheDocument()
+
+    act(() => {
+      agentListener?.({
+        requestId: request.requestId,
+        type: 'subagent',
+        childTaskId: '00000000-0000-4000-8000-000000000611',
+        expertId: '00000000-0000-4000-8000-000000000711',
+        expertName: 'explorer',
+        routingMode: 'native',
+        runtimeCallId: 'call-task-1',
+        state: 'running',
+        reason: 'Review application architecture'
+      })
+      agentListener?.({
+        requestId: request.requestId,
+        type: 'text',
+        delta: '父 Agent 最终回复'
+      })
+    })
+
+    const region = await screen.findByLabelText('子代理状态')
+    expect(within(region).getByText('explorer')).toBeInTheDocument()
+    expect(within(region).getByText('OpenCode 原生'))
+      .toBeInTheDocument()
+    expect(
+      screen.queryByRole('region', {
+        name: '工具执行，共 1 项'
+      })
+    ).not.toBeInTheDocument()
+    const assistantArticle = screen
+      .getByText('父 Agent 最终回复')
+      .closest('article')
+    const orderedBlocks = [
+      ...assistantArticle!.querySelectorAll('.message-blocks > *')
+    ].map((element) => element.textContent)
+    expect(orderedBlocks).toEqual([
+      expect.stringContaining('Review application architecture'),
+      expect.stringContaining('等待子代理结果'),
+      expect.stringContaining('父 Agent 最终回复')
+    ])
+    fireEvent.click(screen.getByText('运行记录'))
+    expect(await screen.findAllByText('子专家')).toHaveLength(1)
+    expect(
+      screen.queryByText(
+        '远端 Runtime 工具：Review application architecture'
+      )
+    ).not.toBeInTheDocument()
+  })
+
+  it('fails an unterminated subagent when the parent reports done', async () => {
+    render(<App />)
+    fireEvent.change(await screen.findByLabelText('向 GoodBuddy 提问'), {
+      target: { value: '检查未完成子任务' }
+    })
+    fireEvent.click(screen.getByLabelText('发送'))
+    await waitFor(() => expect(run).toHaveBeenCalledOnce())
+    const request = run.mock.calls[0]?.[0]
+    if (!request) {
+      throw new Error('Missing request')
+    }
+
+    act(() => {
+      agentListener?.({
+        requestId: request.requestId,
+        type: 'subagent',
+        childTaskId: '00000000-0000-4000-8000-000000000612',
+        expertId: '00000000-0000-4000-8000-000000000712',
+        expertName: 'explorer',
+        routingMode: 'native',
+        runtimeCallId: 'call-task-2',
+        state: 'running',
+        reason: '检查未完成路径'
+      })
+      agentListener?.({
+        requestId: request.requestId,
+        type: 'done'
+      })
+    })
+
+    const region = await screen.findByLabelText('子代理状态')
+    expect(within(region).getByText('失败')).toBeInTheDocument()
+    const subagentCard = within(region).getByRole('group')
+    fireEvent.click(
+      within(subagentCard).getByText('explorer', {
+        selector: 'strong'
+      })
+    )
+    fireEvent(subagentCard, new Event('toggle'))
+    expect(
+      within(region).getByText(
+        '父请求已结束，但子 Agent 未报告完成状态。'
+      )
+    ).toBeInTheDocument()
+    await waitFor(() => {
+      const persistedMessages = vi
+        .mocked(api.conversations.saveLocal)
+        .mock.calls.flatMap(([batch]) =>
+          batch.flatMap((conversation) => conversation.messages)
+        )
+      expect(
+        persistedMessages.some(
+          (message) =>
+            message.subagents?.[0]?.state === 'failed' &&
+            message.subagents[0].error ===
+              '父请求已结束，但子 Agent 未报告完成状态。'
+        )
+      ).toBe(true)
+    })
+  })
+
   it('offers once, session, permanent, and deny for a tool call', async () => {
     render(<App />)
 

src/renderer/src/App.tsx

@@ -128,6 +128,7 @@ import {
   conversationMessageBlocksSchema,
   conversationSubagentActivitySchema,
   interactiveWorkModes,
+  maximumConversationSubagentActivities,
   normalizeInteractiveWorkMode,
   projectChannelLabels
 } from '../../shared/assistant-contracts'
@@ -619,9 +620,57 @@ function upsertMessageToolBlock(
   ]
 }
 
+function upsertMessageSubagentBlock(
+  blocks: ConversationMessageBlock[] | undefined,
+  childTaskId: string,
+  runtimeCallId?: string
+): ConversationMessageBlock[] | undefined {
+  if (!blocks) {
+    return blocks
+  }
+  if (
+    blocks.some(
+      (block) =>
+        block.type === 'subagent' &&
+        block.childTaskId === childTaskId
+    )
+  ) {
+    return blocks
+  }
+  const provisionalIndex = runtimeCallId
+    ? blocks.findIndex(
+        (block) =>
+          block.type === 'tool' &&
+          block.tool.callId === runtimeCallId
+      )
+    : -1
+  if (provisionalIndex >= 0) {
+    return blocks.map((block, index) =>
+      index === provisionalIndex
+        ? {
+            id: block.id,
+            type: 'subagent' as const,
+            childTaskId
+          }
+        : block
+    )
+  }
+  if (blocks.length >= maxMessageBlocks) {
+    return blocks
+  }
+  return [
+    ...blocks,
+    {
+      id: crypto.randomUUID(),
+      type: 'subagent',
+      childTaskId
+    }
+  ]
+}
+
 function terminalizeMessageToolBlocks(
   blocks: ConversationMessageBlock[] | undefined,
-  state: 'failed' | 'cancelled'
+  state: 'failed' | 'cancelled' | 'interrupted'
 ): ConversationMessageBlock[] | undefined {
   return blocks?.map((block) =>
     block.type === 'tool' &&
@@ -1212,7 +1261,8 @@ function isConversation(value: unknown): value is Conversation {
             ))) &&
         (entry.subagents === undefined ||
           (Array.isArray(entry.subagents) &&
-            entry.subagents.length <= 3 &&
+            entry.subagents.length <=
+              maximumConversationSubagentActivities &&
             entry.subagents.every(
               (subagent) =>
                 conversationSubagentActivitySchema.safeParse(subagent)
@@ -4183,7 +4233,10 @@ function App(): React.JSX.Element {
             conversationId: run.conversationId,
             parentTaskId: event.requestId,
             expertId: event.expertId,
-            routingMode: event.routingMode,
+            routingMode:
+              event.routingMode === 'native'
+                ? undefined
+                : event.routingMode,
             title: event.expertName,
             instructions:
               event.reason ??
@@ -4205,8 +4258,23 @@ function App(): React.JSX.Element {
             ? current.map((task) =>
                 task.id === event.childTaskId ? childTask : task
               )
-            : [...current, childTask].slice(0, 100)
+            : [...current, childTask].slice(
+                0,
+                maximumConversationSubagentActivities
+              )
         })
+        if (event.runtimeCallId) {
+          setActivityRecords((current) =>
+            current.filter(
+              (record) =>
+                !(
+                  record.requestId === event.requestId &&
+                  record.kind === 'tool' &&
+                  record.callId === event.runtimeCallId
+                )
+            )
+          )
+        }
         recordActivity({
           conversationId: run.conversationId,
           requestId: event.requestId,
@@ -4216,7 +4284,9 @@ function App(): React.JSX.Element {
           detail: [
             event.routingMode === 'smart'
               ? tRef.current('chat.subagents.smart')
-              : tRef.current('chat.subagents.manual'),
+              : event.routingMode === 'native'
+                ? tRef.current('chat.subagents.native')
+                : tRef.current('chat.subagents.manual'),
             event.reason,
             event.error
           ]
@@ -4239,6 +4309,7 @@ function App(): React.JSX.Element {
             expertId: event.expertId,
             expertName: event.expertName,
             routingMode: event.routingMode,
+            runtimeCallId: event.runtimeCallId,
             state: event.state,
             reason: event.reason,
             output: event.output,
@@ -4246,10 +4317,27 @@ function App(): React.JSX.Element {
           }
           if (index >= 0) {
             subagents[index] = subagent
-          } else if (subagents.length < 3) {
+          } else if (
+            subagents.length <
+            maximumConversationSubagentActivities
+          ) {
             subagents.push(subagent)
           }
-          return { ...message, subagents }
+          const runtimeCallId = event.runtimeCallId
+          return {
+            ...message,
+            subagents,
+            tools: runtimeCallId
+              ? message.tools?.filter(
+                  (tool) => tool.callId !== runtimeCallId
+                )
+              : message.tools,
+            blocks: upsertMessageSubagentBlock(
+              message.blocks,
+              event.childTaskId,
+              runtimeCallId
+            )
+          }
         })
       } else if (event.type === 'approval') {
         recordActivity({
@@ -4342,6 +4430,12 @@ function App(): React.JSX.Element {
               ? 'cancelled'
               : 'failed'
             : 'completed'
+        const incompleteSubagentError = tRef.current(
+          'chat.subagents.incomplete'
+        )
+        const incompleteActivityDetail = tRef.current(
+          'chat.status.activityIncomplete'
+        )
         updateRequestActivity(
           event.requestId,
           terminalStatus,
@@ -4349,25 +4443,27 @@ function App(): React.JSX.Element {
             ? event.message
             : tRef.current('chat.status.taskCompleted')
         )
-        if (event.type === 'error') {
+        setActivityRecords((current) =>
-          setActivityRecords((current) =>
+          current.map((record) =>
-            current.map((record) =>
+            record.requestId === event.requestId &&
-              record.requestId === event.requestId &&
+            record.kind !== 'request' &&
-              record.kind !== 'request' &&
+            (record.status === 'pending' ||
-              (record.status === 'pending' ||
+              record.status === 'running')
-                record.status === 'running')
+              ? {
-                ? {
+                  ...record,
-                    ...record,
+                  status:
-                    status: terminalStatus,
+                    event.type === 'done'
-                    detail: `${record.detail}\n${event.message}`.slice(
+                      ? 'interrupted'
-                      0,
+                      : terminalStatus,
-                      4_000
+                  detail: `${record.detail}\n${
-                    )
+                    event.type === 'done'
-                  }
+                      ? incompleteActivityDetail
-                : record
+                      : event.message
-            )
+                  }`.slice(0, 4_000)
+                }
+              : record
           )
-        }
+        )
         recordActivity({
           conversationId: run.conversationId,
           requestId: event.requestId,
@@ -4382,6 +4478,25 @@ function App(): React.JSX.Element {
               : tRef.current('chat.status.runtimeCompleted'),
           status: terminalStatus
         })
+        setAssistantTasks((current) =>
+          current.map((task) =>
+            task.parentTaskId === event.requestId &&
+            (task.status === 'queued' || task.status === 'running')
+              ? {
+                  ...task,
+                  status:
+                    event.type === 'done'
+                      ? 'failed'
+                      : terminalStatus,
+                  completedAt: new Date().toISOString(),
+                  error:
+                    event.type === 'error'
+                      ? event.message
+                      : incompleteSubagentError
+                }
+              : task
+          )
+        )
         updateMessage(run.conversationId, run.messageId, (message) => {
           const representedToolError =
             event.type === 'error' &&
@@ -4394,7 +4509,7 @@ function App(): React.JSX.Element {
               ? event.status === 'cancelled'
                 ? ('cancelled' as const)
                 : ('failed' as const)
-              : undefined
+              : ('interrupted' as const)
           const fallbackError =
             event.type === 'error' &&
             !representedToolError &&
@@ -4436,6 +4551,26 @@ function App(): React.JSX.Element {
                     : tool
                 )
               : message.tools,
+            subagents: message.subagents?.map((subagent) =>
+              subagent.state === 'queued' ||
+              subagent.state === 'running'
+                ? {
+                    ...subagent,
+                    state:
+                      event.type === 'error'
+                        ? event.status === 'cancelled'
+                          ? ('cancelled' as const)
+                          : ('failed' as const)
+                        : ('failed' as const),
+                    ...(event.type === 'error' &&
+                    event.status !== 'cancelled'
+                      ? { error: event.message.slice(0, 1_000) }
+                      : event.type === 'done'
+                        ? { error: incompleteSubagentError }
+                        : {})
+                  }
+                : subagent
+            ),
             blocks: toolTerminalState
               ? terminalizeMessageToolBlocks(
                   appendMessageContentBlock(

src/renderer/src/ChatTimeline.test.tsx

@@ -246,26 +246,110 @@ describe('ChatTimeline', () => {
       />
     )
 
-    const region = screen.getByLabelText('子专家状态')
+    const region = screen.getByLabelText('子代理状态')
     const summary = screen.getByText('综合结果').parentElement
     const messageBody = region.parentElement
     expect(summary?.parentElement).toBe(messageBody)
     expect(
       Array.from(messageBody?.children ?? []).indexOf(region)
-    ).toBeLessThan(
+    ).toBeGreaterThan(
       Array.from(messageBody?.children ?? []).indexOf(summary!)
     )
     const cards = within(region).getAllByRole('group')
     expect(cards).toHaveLength(3)
+    const experts = ['研究专家', '代码专家', '安全专家']
     for (const [index, output] of [
       '研究专家的独立结论',
       '代码专家的独立结论',
       '安全专家的独立结论'
     ].entries()) {
-      expect(within(cards[index]!).getByText(output)).toBeInTheDocument()
+      expect(
+        within(cards[index]!).queryByText(output)
+      ).not.toBeInTheDocument()
       expect(cards[index]).not.toHaveAttribute('open')
-      fireEvent.click(within(cards[index]!).getByText(/专家$/u))
+      fireEvent.click(
+        within(cards[index]!).getByText(experts[index]!, {
+          selector: 'strong'
+        })
+      )
+      fireEvent(cards[index]!, new Event('toggle'))
       expect(cards[index]).toHaveAttribute('open')
+      expect(within(cards[index]!).getByText(output)).toBeInTheDocument()
     }
   })
+
+  it('labels native OpenCode subagents and shows their task description', () => {
+    const messages: Message[] = [
+      {
+        id: 'assistant-message',
+        role: 'assistant',
+        content: '',
+        createdAt: 1_775_000_000_000,
+        state: 'streaming',
+        blocks: [
+          {
+            id: '00000000-0000-4000-8000-000000000121',
+            type: 'reasoning',
+            content: '先分析桌面目录'
+          },
+          {
+            id: '00000000-0000-4000-8000-000000000122',
+            type: 'subagent',
+            childTaskId:
+              '00000000-0000-4000-8000-000000000111'
+          },
+          {
+            id: '00000000-0000-4000-8000-000000000123',
+            type: 'text',
+            content: '父 Agent 最终回复'
+          }
+        ],
+        subagents: [
+          {
+            childTaskId: '00000000-0000-4000-8000-000000000111',
+            expertId: '00000000-0000-4000-8000-000000000211',
+            expertName: 'explorer',
+            routingMode: 'native',
+            runtimeCallId: 'call-task-1',
+            state: 'running',
+            reason: 'Review application architecture'
+          }
+        ]
+      }
+    ]
+
+    render(
+      <ChatTimeline
+        artifactById={new Map()}
+        conversationId="conversation-1"
+        hiddenMessageCount={0}
+        isUnusedConversation={false}
+        locale="zh-CN"
+        messageStartIndex={0}
+        messages={messages}
+        {...callbacks}
+        retryContent=""
+        totalMessageCount={messages.length}
+      />
+    )
+
+    expect(screen.getByText('子代理')).toBeInTheDocument()
+    expect(screen.getByText('OpenCode 原生')).toBeInTheDocument()
+    expect(screen.getByText('任务')).toBeInTheDocument()
+    expect(screen.getByText('进行中')).toBeInTheDocument()
+    expect(
+      screen.getByText('Review application architecture')
+    ).toBeInTheDocument()
+    const assistantArticle = screen
+      .getByText('父 Agent 最终回复')
+      .closest('article')
+    const orderedBlocks = [
+      ...assistantArticle!.querySelectorAll('.message-blocks > *')
+    ].map((element) => element.textContent)
+    expect(orderedBlocks).toEqual([
+      expect.stringContaining('先分析桌面目录'),
+      expect.stringContaining('Review application architecture'),
+      expect.stringContaining('父 Agent 最终回复')
+    ])
+  })
 })

src/renderer/src/ChatTimeline.tsx

@@ -1,14 +1,19 @@
 import {
   Bot,
+  CheckCircle2,
+  ChevronRight,
+  Clock3,
   Download,
   FileText,
   Library,
   ClockFading,
+  LoaderCircle,
   ShieldCheck,
   TerminalSquare,
-  UserRound
+  UserRound,
+  XCircle
 } from 'lucide-react'
-import { memo, useEffect, useRef } from 'react'
+import { memo, useEffect, useRef, useState } from 'react'
 import { useTranslation } from 'react-i18next'
 import type {
   ApprovalDecision,
@@ -78,19 +83,40 @@ export type ImageViewerItem = {
 type MessageBlockRenderItem =
   | {
       kind: 'block'
-      block: Exclude<ConversationMessageBlock, { type: 'tool' }>
+      block: Extract<
+        ConversationMessageBlock,
+        { type: 'text' | 'reasoning' }
+      >
     }
   | {
       kind: 'tools'
       id: string
       tools: ToolActivity[]
     }
+  | {
+      kind: 'subagents'
+      id: string
+      childTaskIds: string[]
+    }
 
 function groupMessageBlocks(
   blocks: ConversationMessageBlock[]
 ): MessageBlockRenderItem[] {
   const items: MessageBlockRenderItem[] = []
   for (const block of blocks) {
+    if (block.type === 'subagent') {
+      const previous = items.at(-1)
+      if (previous?.kind === 'subagents') {
+        previous.childTaskIds.push(block.childTaskId)
+      } else {
+        items.push({
+          kind: 'subagents',
+          id: block.id,
+          childTaskIds: [block.childTaskId]
+        })
+      }
+      continue
+    }
     if (block.type !== 'tool') {
       items.push({ kind: 'block', block })
       continue
@@ -219,6 +245,90 @@ function ToolExecutionList({
   )
 }
 
+const SubagentStatusCard = memo(function SubagentStatusCard({
+  subagent
+}: {
+  subagent: SubagentActivity
+}): React.JSX.Element {
+  const { t } = useTranslation('app')
+  const [expanded, setExpanded] = useState(false)
+  const StateIcon =
+    subagent.state === 'completed'
+      ? CheckCircle2
+      : subagent.state === 'running'
+        ? LoaderCircle
+        : subagent.state === 'queued'
+          ? Clock3
+          : XCircle
+  const source =
+    subagent.routingMode === 'smart'
+      ? t('chat.subagents.smart')
+      : subagent.routingMode === 'native'
+        ? t('chat.subagents.native')
+        : t('chat.subagents.manual')
+
+  return (
+    <details
+      className={`subagent-status-card subagent-status-card--${subagent.state}`}
+      onToggle={(event) => setExpanded(event.currentTarget.open)}
+    >
+      <summary>
+        <span className="subagent-status-card__identity">
+          <span className="subagent-status-card__heading">
+            <span className="subagent-status-card__badge">
+              <Bot aria-hidden="true" size={13} />
+              {t('chat.subagents.badge')}
+            </span>
+            <strong>{subagent.expertName}</strong>
+          </span>
+          {subagent.reason && (
+            <span className="subagent-status-card__task">
+              <small>{t('chat.subagents.task')}</small>
+              <span>{subagent.reason}</span>
+            </span>
+          )}
+          <small className="subagent-status-card__source">
+            {source}
+          </small>
+        </span>
+        <span
+          className={`subagent-status-card__status subagent-status-card__status--${subagent.state}`}
+        >
+          <StateIcon aria-hidden="true" size={14} />
+          {t(`chat.subagents.states.${subagent.state}`)}
+        </span>
+        <ChevronRight
+          aria-hidden="true"
+          className="subagent-status-card__chevron"
+          size={15}
+        />
+      </summary>
+      {expanded && (
+        <div className="subagent-status-card__details">
+          {subagent.output ? (
+            <section>
+              <strong>{t('chat.subagents.output')}</strong>
+              <div className="markdown-content">
+                <MarkdownRenderer>{subagent.output}</MarkdownRenderer>
+              </div>
+            </section>
+          ) : (
+            <p>{t('chat.subagents.noOutput')}</p>
+          )}
+          {subagent.error &&
+            (subagent.state === 'failed' ||
+              subagent.state === 'cancelled') && (
+              <section className="subagent-status-card__error">
+                <strong>{t('chat.subagents.error')}</strong>
+                <p>{subagent.error}</p>
+              </section>
+            )}
+        </div>
+      )}
+    </details>
+  )
+})
+
 function SubagentStatusList({
   subagents
 }: {
@@ -231,44 +341,11 @@ function SubagentStatusList({
       aria-label={t('chat.subagents.region')}
       className="subagent-status-list"
     >
-      {subagents.slice(0, 3).map((subagent) => (
+      {subagents.map((subagent) => (
-        <details
+        <SubagentStatusCard
-          className={`subagent-status-card subagent-status-card--${subagent.state}`}
           key={subagent.childTaskId}
-        >
+          subagent={subagent}
-          <summary>
+        />
-            <Bot aria-hidden="true" size={15} />
-            <span className="subagent-status-card__identity">
-              <strong>{subagent.expertName}</strong>
-              <small>
-                {subagent.routingMode === 'smart'
-                  ? t('chat.subagents.smart')
-                  : t('chat.subagents.manual')}
-              </small>
-            </span>
-            <span>{t(`chat.subagents.states.${subagent.state}`)}</span>
-          </summary>
-          <div className="subagent-status-card__details">
-            {subagent.output ? (
-              <section>
-                <strong>{t('chat.subagents.output')}</strong>
-                <div className="markdown-content">
-                  <MarkdownRenderer>{subagent.output}</MarkdownRenderer>
-                </div>
-              </section>
-            ) : (
-              <p>{t('chat.subagents.noOutput')}</p>
-            )}
-            {(subagent.error || subagent.reason) &&
-              (subagent.state === 'failed' ||
-                subagent.state === 'cancelled') && (
-                <section className="subagent-status-card__error">
-                  <strong>{t('chat.subagents.error')}</strong>
-                  <p>{subagent.error ?? subagent.reason}</p>
-                </section>
-              )}
-          </div>
-        </details>
       ))}
     </section>
   )
@@ -327,6 +404,20 @@ function ChatMessageRowView({
   const compressionMarkers =
     message.contextCompressions ??
     (message.contextCompression ? [message.contextCompression] : [])
+  const subagentsById = new Map(
+    (message.subagents ?? []).map((subagent) => [
+      subagent.childTaskId,
+      subagent
+    ])
+  )
+  const orderedSubagentIds = new Set(
+    message.blocks
+      ?.filter((block) => block.type === 'subagent')
+      .map((block) => block.childTaskId)
+  )
+  const unorderedSubagents = message.subagents?.filter(
+    (subagent) => !orderedSubagentIds.has(subagent.childTaskId)
+  )
   const compressionLabel = (
     compression: ConversationContextCompressionMarker
   ): string =>
@@ -468,14 +559,19 @@ function ChatMessageRowView({
             })}
           </div>
         )}
-        {message.subagents && message.subagents.length > 0 && (
-          <SubagentStatusList subagents={message.subagents} />
-        )}
         {message.blocks && message.blocks.length > 0 ? (
           <div className="message-blocks">
             {groupMessageBlocks(message.blocks).map((item) =>
               item.kind === 'tools' ? (
                 <ToolExecutionList key={item.id} tools={item.tools} />
+              ) : item.kind === 'subagents' ? (
+                <SubagentStatusList
+                  key={item.id}
+                  subagents={item.childTaskIds.flatMap((childTaskId) => {
+                    const subagent = subagentsById.get(childTaskId)
+                    return subagent ? [subagent] : []
+                  })}
+                />
               ) : item.block.type === 'reasoning' ? (
                 <MessageReasoning
                   content={item.block.content}
@@ -510,6 +606,9 @@ function ChatMessageRowView({
             )}
           </>
         )}
+        {unorderedSubagents && unorderedSubagents.length > 0 && (
+          <SubagentStatusList subagents={unorderedSubagents} />
+        )}
         {message.artifactIds?.map((artifactId) => {
           const candidate = artifactById.get(artifactId)
           const artifact =

src/renderer/src/MagicNotesWorkspace.test.tsx

@@ -399,6 +399,11 @@ describe('MagicNotesWorkspace', () => {
     expect(
       screen.getByRole('button', { name: /发布笔记/ })
     ).toHaveAttribute('aria-pressed', 'true')
+    const updatedTime = screen
+      .getByRole('button', { name: /发布笔记/ })
+      .querySelector('time')
+    expect(updatedTime).toHaveAttribute('datetime', detail.updatedAt)
+    expect(updatedTime).toHaveTextContent('更新于')
     fireEvent.click(screen.getByRole('tab', { name: '待办' }))
     expect(screen.getByText('准备演示')).toBeInTheDocument()
     const selectedTodoButton = screen

src/renderer/src/MagicNotesWorkspace.tsx

@@ -1657,12 +1657,21 @@ export function MagicNotesWorkspace({
                     {note.preview || t('notes.noPreview')}
                   </span>
                   <span className="magic-note-list-item__meta">
-                    {t(
+                    <span>
-                      note.entryCount === 1
+                      {t(
-                        ? 'notes.entryCountOne'
+                        note.entryCount === 1
-                        : 'notes.entryCountOther',
+                          ? 'notes.entryCountOne'
-                      { count: note.entryCount }
+                          : 'notes.entryCountOther',
-                    )}
+                        { count: note.entryCount }
+                      )}
+                    </span>
+                    <time dateTime={note.updatedAt}>
+                      {t('notes.updatedAt', {
+                        date: dateFormatter.format(
+                          new Date(note.updatedAt)
+                        )
+                      })}
+                    </time>
                   </span>
                 </button>
               ))

src/renderer/src/i18n/locales/en-US/app.ts

@@ -322,6 +322,8 @@ export const app = {
       taskCompleted: 'Task completed',
       taskFailed: 'Task failed',
       runtimeCompleted: 'Agent Runtime completed its response',
+      activityIncomplete:
+        'The parent request ended before this activity reported completion',
       answerSubmitted: 'Answer submitted; OpenCode is continuing',
       questionSkipped: 'Question skipped'
     },
@@ -345,12 +347,17 @@ export const app = {
     },
     subagents: {
       region: 'Subagent status',
-      smart: 'Smart routing',
+      badge: 'Subagent',
-      manual: 'Selected manually',
+      smart: 'GoodBuddy smart routing',
+      manual: 'GoodBuddy selected expert',
+      native: 'Native OpenCode',
       fallbackTask: '{{name}} subagent task',
-      output: 'Expert output',
+      task: 'Task',
+      output: 'Subagent output',
       error: 'Execution details',
-      noOutput: 'This expert has no output to display yet.',
+      noOutput: 'This subagent has no output to display yet.',
+      incomplete:
+        'The parent request ended before the subagent reported completion.',
       states: {
         queued: 'Queued',
         running: 'Running',

src/renderer/src/i18n/locales/en-US/magicNotes.ts

@@ -82,6 +82,7 @@ export const magicNotes = {
     noPreview: 'No entries yet',
     entryCountOne: '{{count}} entry',
     entryCountOther: '{{count}} entries',
+    updatedAt: 'Updated {{date}}',
     emptySelectionTitle: 'No note selected',
     emptySelectionDescription:
       'Select a note on the left, or create a note to start writing.',

src/renderer/src/i18n/locales/en-US/settings.ts

@@ -742,7 +742,7 @@ export const settings = {
       incompatibleImageProtocol:
         'Incompatible (image generation protocol is unsupported)',
       incompatibleHarnessProtocol:
-        'Incompatible (requires Chat Completions, an API key, and a secure endpoint)',
+        'Incompatible (requires Chat Completions, an API key, and an HTTP/HTTPS endpoint)',
       secureStorageWarning:
         'Secure system key storage is unavailable. Use an environment variable to avoid storing an API Key in plaintext.'
     },

src/renderer/src/i18n/locales/zh-CN/app.ts

@@ -308,6 +308,7 @@ export const app = {
       taskCompleted: '任务执行完成',
       taskFailed: '任务执行失败',
       runtimeCompleted: 'Agent Runtime 已完成响应',
+      activityIncomplete: '父请求已结束，但该活动未报告完成状态',
       answerSubmitted: '回答已提交，OpenCode 正在继续执行',
       questionSkipped: '已跳过问题'
     },
@@ -330,13 +331,17 @@ export const app = {
       }
     },
     subagents: {
-      region: '子专家状态',
+      region: '子代理状态',
-      smart: '智能路由',
+      badge: '子代理',
-      manual: '手动指定',
+      smart: 'GoodBuddy 智能路由',
+      manual: 'GoodBuddy 指定专家',
+      native: 'OpenCode 原生',
       fallbackTask: '{{name}} 子专家任务',
-      output: '专家输出',
+      task: '任务',
+      output: '子代理输出',
       error: '执行说明',
-      noOutput: '该专家暂时没有可显示的输出。',
+      noOutput: '子代理暂时没有可显示的输出。',
+      incomplete: '父请求已结束，但子 Agent 未报告完成状态。',
       states: {
         queued: '等待中',
         running: '进行中',

src/renderer/src/i18n/locales/zh-CN/magicNotes.ts

@@ -77,6 +77,7 @@ export const magicNotes = {
     noPreview: '还没有记录',
     entryCountOne: '{{count}} 条记录',
     entryCountOther: '{{count}} 条记录',
+    updatedAt: '更新于 {{date}}',
     emptySelectionTitle: '还没有选择笔记',
     emptySelectionDescription: '从左侧选择笔记，或新建一篇笔记开始记录。',
     newEntryLabel: '新记录内容',

src/renderer/src/i18n/locales/zh-CN/settings.ts

@@ -681,7 +681,7 @@ export const settings = {
       incompatible: '不兼容',
       incompatibleImageProtocol: '不兼容（不支持图像生成协议）',
       incompatibleHarnessProtocol:
-        '不兼容（需 Chat Completions、API Key 和安全地址）',
+        '不兼容（需 Chat Completions、API Key 和 HTTP/HTTPS 地址）',
       secureStorageWarning:
         '当前系统密钥服务不可用。为了避免明文落盘，请使用环境变量提供 API Key。'
     },

src/renderer/src/styles.css

@@ -1609,8 +1609,24 @@
 }
 
 .magic-note-list-item__meta {
+  display: flex;
+  min-width: 0;
+  align-items: center;
+  justify-content: space-between;
   color: var(--text-muted);
   font-size: var(--font-caption);
+  gap: var(--space-2);
+}
+
+.magic-note-list-item__meta > :where(span, time) {
+  overflow: hidden;
+  text-overflow: ellipsis;
+  white-space: nowrap;
+}
+
+.magic-note-list-item__meta time {
+  flex-shrink: 1;
+  text-align: right;
 }
 
 .magic-notes-stream-pane {
@@ -4891,7 +4907,8 @@ button > svg {
 }
 
 .message-blocks .message-reasoning,
-.message-blocks .tool-execution-list {
+.message-blocks .tool-execution-list,
+.message-blocks .subagent-status-list {
   margin: 0;
 }
 
@@ -5243,25 +5260,25 @@ button > svg {
 
 .subagent-status-list {
   display: grid;
-  margin-top: var(--space-2);
+  margin: var(--space-2) 0 var(--space-4);
   gap: var(--space-2);
 }
 
 .subagent-status-card {
   border: 1px solid var(--border-default);
-  border-radius: var(--radius-control);
+  border-radius: var(--radius-card);
-  background: var(--surface-subtle);
+  background: var(--surface-raised);
   color: var(--text-secondary);
   overflow: hidden;
 }
 
 .subagent-status-card > summary {
   display: grid;
-  align-items: center;
+  align-items: start;
-  padding: var(--space-2) var(--space-3);
+  padding: var(--space-3) var(--space-4);
   cursor: pointer;
-  gap: var(--space-2);
+  gap: var(--space-3);
-  grid-template-columns: auto minmax(0, 1fr) auto;
+  grid-template-columns: minmax(0, 1fr) auto auto;
   list-style: none;
 }
 
@@ -5278,38 +5295,130 @@ button > svg {
   outline-offset: -2px;
 }
 
+.subagent-status-card--queued {
+  box-shadow: inset 3px 0 0 var(--warning);
+}
+
 .subagent-status-card--running {
-  border-color: var(--accent-selected);
+  box-shadow: inset 3px 0 0 var(--accent);
+}
+
+.subagent-status-card--completed {
+  box-shadow: inset 3px 0 0 var(--success);
 }
 
 .subagent-status-card--failed,
 .subagent-status-card--cancelled {
   border-color: var(--danger-border);
   background: var(--danger-subtle);
+  box-shadow: inset 3px 0 0 var(--danger);
 }
 
 .subagent-status-card__identity {
   display: grid;
   min-width: 0;
+  gap: var(--space-2);
+}
+
+.subagent-status-card__heading {
+  display: flex;
+  min-width: 0;
+  align-items: center;
+  flex-wrap: wrap;
+  gap: var(--space-2);
+}
+
+.subagent-status-card__badge {
+  display: inline-flex;
+  min-height: 22px;
+  padding: 2px var(--space-2);
+  align-items: center;
+  border: 1px solid var(--accent-selected);
+  border-radius: var(--radius-control);
+  background: var(--accent-subtle);
+  color: var(--accent);
+  font-size: var(--font-caption);
+  font-weight: 700;
   gap: var(--space-1);
 }
 
-.subagent-status-card summary strong,
+.subagent-status-card__heading > strong {
-.subagent-status-card summary > span:last-child {
   color: var(--text-primary);
-  font-size: var(--font-body);
+  font-size: var(--font-section-title);
+  overflow-wrap: anywhere;
 }
 
-.subagent-status-card summary small {
+.subagent-status-card__task {
+  display: grid;
+  min-width: 0;
+  gap: var(--space-1);
+}
+
+.subagent-status-card__task > small,
+.subagent-status-card__source {
   color: var(--text-muted);
   font-size: var(--font-caption);
 }
 
+.subagent-status-card__task > span {
+  color: var(--text-primary);
+  font-size: var(--font-body);
+  line-height: 1.5;
+  overflow-wrap: anywhere;
+}
+
+.subagent-status-card__status {
+  display: inline-flex;
+  min-height: 24px;
+  padding: 2px var(--space-2);
+  align-items: center;
+  border-radius: var(--radius-control);
+  font-size: var(--font-caption);
+  font-weight: 700;
+  gap: var(--space-1);
+  white-space: nowrap;
+}
+
+.subagent-status-card__status--queued {
+  background: var(--warning-subtle);
+  color: var(--warning);
+}
+
+.subagent-status-card__status--running {
+  background: var(--accent-subtle);
+  color: var(--accent);
+}
+
+.subagent-status-card__status--completed {
+  background: var(--success-subtle);
+  color: var(--success);
+}
+
+.subagent-status-card__status--failed,
+.subagent-status-card__status--cancelled {
+  background: var(--danger-subtle);
+  color: var(--danger);
+}
+
+.subagent-status-card__status--running > svg {
+  animation: subagent-status-spin 900ms linear infinite;
+}
+
+.subagent-status-card__chevron {
+  margin-top: var(--space-1);
+  color: var(--text-muted);
+  transition: transform var(--motion-fast) ease-out;
+}
+
+.subagent-status-card[open] .subagent-status-card__chevron {
+  transform: rotate(90deg);
+}
+
 .subagent-status-card__details {
   display: grid;
-  padding: var(--space-3);
+  padding: var(--space-3) var(--space-4) var(--space-4);
   border-top: 1px solid var(--border-subtle);
-  background: var(--surface-raised);
+  background: var(--surface-subtle);
   gap: var(--space-3);
 }
 
@@ -5333,6 +5442,12 @@ button > svg {
   color: var(--danger);
 }
 
+@keyframes subagent-status-spin {
+  to {
+    transform: rotate(360deg);
+  }
+}
+
 .approval-card {
   display: grid;
   align-items: center;

src/shared/assistant-contracts.test.ts

@@ -242,4 +242,49 @@ describe('conversation persistence contracts', () => {
 
     expect(localConversationSaveBatchSchema.parse(batch)).toEqual(batch)
   })
+
+  it('accepts persisted native OpenCode subagent activity', () => {
+    const batch = [
+      {
+        header: {
+          id: '00000000-0000-4000-8000-000000000311',
+          title: 'OpenCode 子 Agent',
+          updatedAt: 1
+        },
+        messages: [
+          {
+            id: '00000000-0000-4000-8000-000000000312',
+            role: 'assistant',
+            content: '',
+            createdAt: 1,
+            state: 'complete',
+            blocks: [
+              {
+                id: '00000000-0000-4000-8000-000000000315',
+                type: 'subagent',
+                childTaskId:
+                  '00000000-0000-4000-8000-000000000313'
+              }
+            ],
+            subagents: [
+              {
+                childTaskId:
+                  '00000000-0000-4000-8000-000000000313',
+                expertId:
+                  '00000000-0000-4000-8000-000000000314',
+                expertName: 'explorer',
+                routingMode: 'native',
+                runtimeCallId: 'call-task-1',
+                state: 'completed',
+                reason: 'Review application architecture',
+                output: 'Architecture review complete.'
+              }
+            ]
+          }
+        ]
+      }
+    ]
+
+    expect(localConversationSaveBatchSchema.parse(batch)).toEqual(batch)
+  })
 })

src/shared/assistant-contracts.ts

@@ -172,6 +172,13 @@ export const conversationMessageBlockSchema = z.discriminatedUnion('type', [
       type: z.literal('tool'),
       tool: conversationToolActivitySchema
     })
+    .strict(),
+  z
+    .object({
+      id: assistantIdSchema,
+      type: z.literal('subagent'),
+      childTaskId: assistantIdSchema
+    })
     .strict()
 ])
 
@@ -186,12 +193,16 @@ export type ConversationMessageBlock = z.infer<
   typeof conversationMessageBlockSchema
 >
 
+export const maximumConversationToolActivities = 100
+export const maximumConversationSubagentActivities = 100
+
 export const conversationSubagentActivitySchema = z
   .object({
     childTaskId: assistantIdSchema,
     expertId: assistantIdSchema,
     expertName: z.string().trim().min(1).max(80),
-    routingMode: z.enum(['manual', 'smart']),
+    routingMode: z.enum(['manual', 'smart', 'native']),
+    runtimeCallId: z.string().trim().min(1).max(256).optional(),
     state: z.enum([
       'queued',
       'running',
@@ -240,10 +251,13 @@ export const conversationMessageSchema = z
       .array(conversationContextCompressionMarkerSchema)
       .max(2)
       .optional(),
-    tools: z.array(conversationToolActivitySchema).max(100).optional(),
+    tools: z
+      .array(conversationToolActivitySchema)
+      .max(maximumConversationToolActivities)
+      .optional(),
     subagents: z
       .array(conversationSubagentActivitySchema)
-      .max(3)
+      .max(maximumConversationSubagentActivities)
       .optional(),
     sources: z.array(z.string().max(8_192)).max(100).optional(),
     sourceReferences: z

src/shared/contracts.ts

@@ -21,6 +21,7 @@ import {
   conversationAttachmentSchema,
   conversationHistoryMessageSchema,
   conversationContextCompressionStateSchema,
+  conversationSubagentActivitySchema,
   legacyWorkModeSchema,
   maximumConversationHistoryCharacters,
   maximumConversationHistoryMessages,
@@ -830,7 +831,7 @@ export const runtimeSettingsInputSchema = z
           code: 'custom',
           path: ['deepseekHarnessModelSource'],
           message:
-            'DeepSeek Harness 仅支持使用 API Key 的安全 OpenAI 兼容 Chat Completions 连接'
+            'DeepSeek Harness 仅支持使用 API Key 的 OpenAI 兼容 Chat Completions 连接'
         })
       }
       const compressionSource = settings.contextCompression?.modelSource
@@ -1171,24 +1172,11 @@ export const approvalDecisionSchema = z.enum([
 
 export type ApprovalDecision = z.infer<typeof approvalDecisionSchema>
 
-export const subagentEventSchema = z
+export const subagentEventSchema =
-  .object({
+  conversationSubagentActivitySchema
+  .extend({
     requestId: z.string().uuid(),
-    type: z.literal('subagent'),
+    type: z.literal('subagent')
-    childTaskId: z.string().uuid(),
-    expertId: z.string().uuid(),
-    expertName: z.string().trim().min(1).max(80),
-    routingMode: z.enum(['manual', 'smart']),
-    state: z.enum([
-      'queued',
-      'running',
-      'completed',
-      'failed',
-      'cancelled'
-    ]),
-    reason: z.string().trim().min(1).max(240).optional(),
-    output: z.string().optional(),
-    error: z.string().trim().min(1).max(1_000).optional()
   })
   .strict()
 

src/shared/deepseek-harness-compatibility.test.ts

@@ -9,9 +9,14 @@ describe('DeepSeek Harness model compatibility', () => {
     'https://api.deepseek.com',
     'https://gateway.example/openai/v1',
     'https://gateway.example:8443/v1',
+    'http://gateway.example/v1',
+    'http://192.168.1.50:11434/v1',
     'http://127.0.0.1:11434/v1',
     'http://localhost:1234/v1',
-    'http://[::1]:8080/v1'
+    'http://[::1]:8080/v1',
+    'https://gateway.example/v1?api-version=2025-01-01',
+    'https://gateway.example/v1#chat',
+    'http://user:password@gateway.example/v1'
   ])('accepts OpenAI-compatible base URL %s', (baseUrl) => {
     expect(isDeepSeekHarnessCompatibleBaseUrl(baseUrl)).toBe(true)
     expect(
@@ -24,12 +29,10 @@ describe('DeepSeek Harness model compatibility', () => {
   })
 
   it.each([
-    'http://gateway.example/v1',
     'file:///private/config',
-    'https://user:secret@gateway.example/v1',
+    'ftp://gateway.example/v1',
-    'https://gateway.example/v1?api-version=2025-01-01',
+    'not-a-url'
-    'https://gateway.example/v1#chat'
+  ])('rejects unsupported base URL %s', (baseUrl) => {
-  ])('rejects unsafe or unsupported base URL %s', (baseUrl) => {
     expect(isDeepSeekHarnessCompatibleBaseUrl(baseUrl)).toBe(false)
   })
 

src/shared/deepseek-harness-compatibility.ts

@@ -4,35 +4,14 @@ type DeepSeekHarnessModelProfile = {
   authentication: string
 }
 
-function isLoopbackHostname(hostname: string): boolean {
-  const normalized = hostname
-    .toLowerCase()
-    .replace(/^\[|\]$/gu, '')
-  return (
-    normalized === 'localhost' ||
-    normalized === '127.0.0.1' ||
-    normalized === '::1'
-  )
-}
-
 export function isDeepSeekHarnessCompatibleBaseUrl(
   value: string
 ): boolean {
   try {
     const url = new URL(value)
-    if (
-      !url.hostname ||
-      url.username ||
-      url.password ||
-      url.search ||
-      url.hash
-    ) {
-      return false
-    }
     return (
-      url.protocol === 'https:' ||
+      Boolean(url.hostname) &&
-      (url.protocol === 'http:' &&
+      (url.protocol === 'http:' || url.protocol === 'https:')
-        isLoopbackHostname(url.hostname))
     )
   } catch {
     return false

src/shared/remote-agent-contracts.ts

@@ -15,6 +15,9 @@ import {
   modelBridgePolicySchema
 } from './model-bridge-contracts'
 
+export const UNBOUNDED_REMOTE_PROMPT_DEADLINE =
+  '9999-12-31T23:59:59.999Z'
+
 export const runtimeSessionBindingStateSchema = z.enum([
   'opening',
   'ready',