goodbuddy/resources/release-notes.json

{
  "formatVersion": 1,
  "releases": [
    {
      "version": "0.15.7",
      "releasedAt": "2026-10-01",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**减少流式回复期间的重复渲染**：较长回答持续输出时，已完成的 Markdown 块可复用，仅重新解析增长中的尾部，并减少会话侧栏和活动摘要的重复计算。涉及全文语义的内容仍使用完整解析，默认流式刷新节奏保持不变；性能测量来自限定的组件场景，不代表所有设备或完整应用的提速比例。",
            "**运行记录按需刷新**：进入运行记录页面时读取快照，页面可见期间每 15 秒刷新一次，也可使用页头统一刷新按钮更新记录与 Token 用量。页面不再随每个实时事件重新计算，显示可能存在短暂延迟。",
            "**单个自定义 MCP 故障不再拖垮正常工具**：本地 Runtime 发现工具时，失败服务的工具被排除，正常服务仍可使用；动态目录刷新失败会清空该服务的旧工具，避免继续使用失效目录。首次连接失败的服务在下一请求重试，取消与授权撤销继续生效；这不是远程 Agent 的新恢复能力。",
            "**在模型连接中统一选择下载源**：模型类型页签右侧显示当前受管模型下载源，可打开弹窗选择 ModelScope 或 Hugging Face，确认后才保存，取消不改变已保存选择。向量、语音和 OCR 的更换来源入口直接打开同一弹窗；设置只影响后续受管本地模型下载，不会静默换源或修改模型 API、应用更新和 Agent 包来源。",
            "**设备共享测试功能改为显式启用**：先在应用中心启用设备共享，再从应用菜单或应用中心进入；缺失启用设置时默认关闭，已明确保存的选择保留。关闭会退出页面，但不清除目录或撤销已发布记录；仍需另行运行 ShareServer，仅管理元数据，不提供远程执行或知识正文访问。"
          ],
          "notices": [
            "本轮只更新 Desktop，Agent 保持 **0.15.1**，无需重新下载 Agent 包。数据库仍为 schema 52；从较旧版本升级前，请关闭应用并备份完整数据库及相邻 `notes/` 目录，降级需恢复升级前备份。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Reduce repeated rendering during streamed replies**: Completed Markdown blocks can be reused while only the growing tail is parsed, with less repeated work for conversation rows and activity summaries. Content with document-wide semantics still uses full parsing, and the default streaming cadence is unchanged; component benchmarks do not establish a speedup for every device or the complete application.",
            "**Refresh activity records on demand**: Opening the activity page loads a snapshot, refreshes it every 15 seconds while visible, and offers one header button to refresh records and token usage. The page no longer recalculates for every live event, so displayed records may briefly lag behind.",
            "**Keep healthy tools available when one custom MCP service fails**: Local Runtime tool discovery excludes failed services while retaining healthy ones, and a failed dynamic catalog refresh clears that service's stale tools. Initially failed connections are retried on the next request, while cancellation and authorization revocation remain effective; this is not a new remote Agent recovery capability.",
            "**Choose managed model download sources in Model Connections**: The model-type tab row shows the current source and opens a dialog to select ModelScope or Hugging Face, saving only on confirmation and preserving the saved choice on cancel. Embedding, speech, and OCR source-change entries open the same dialog; this affects subsequent managed local model downloads only, without silent source fallback or changes to model APIs, application updates, or Agent package sources.",
            "**Explicitly enable the device-sharing test feature**: Enable Device Sharing in Application Center before opening it from the application menu or manager; a missing preference defaults to off, while explicitly saved choices are preserved. Disabling it exits the page without clearing catalogs or revoking publications; a separately running ShareServer is still required, and this remains metadata management without remote execution or knowledge-content access."
          ],
          "notices": [
            "This is a Desktop-only update; Agent remains **0.15.1**, with no Agent package download required. The database remains at schema 52; before upgrading from older releases, close the app and back up the complete database and adjacent `notes/` directory, and restore a pre-upgrade backup to downgrade."
          ]
        }
      }
    },
    {
      "version": "0.15.6",
      "releasedAt": "2026-10-01",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**可选的透明磨砂外观**：在外观设置中开启“透明磨砂特效”，聊天内容可滚动经过顶栏背后，执行清单浮层也使用磨砂背景。默认关闭，选择会保存并在重启后恢复；效果只作用于应用内部，不透出操作系统桌面。",
            "**设备共享元数据入口**：从应用菜单或应用管理打开“设备共享”，连接另行运行的 ShareServer 服务后，可注册本机、浏览设备和发布目录、发布能力或知识的元数据并撤销本机 ID 的条目。首期面向本地未认证服务，设备 ID 不是所有权认证；不执行远程能力，不搜索、读取或下载知识正文，权限选项仅为元数据声明。"
          ],
          "fixes": [
            "**新建笔记入口保持可见**：“新建笔记”统一放在魔法笔记页面标题栏，不再藏在列表工具栏中。打开详情或待办时仍可发起新建，并沿用未保存草稿保护。",
            "**改善本地 OpenCode 工具连接恢复**：明确的暂时性 MCP 传输故障会在清理成功后最多重试一次，超时或取消后结果不确定的连接名称不再复用，避免晚到初始化覆盖新连接。鉴权、协议和未知错误不会自动重试，Prompt 与工具调用不会重放；该修正不代表远程 Agent 新增恢复能力。",
            "**减少输入和切换会话时的重复处理**：输入时未变化的消息历史及任务条不再反复渲染，切换已缓存会话时复用稳定的消息回调。实时输出、滚动位置、草稿和异步附件归属保持不变，执行统计索引也减少了轮询时对事件记录的扫描。",
            "**草稿和队列图片可直接放大预览**：点击图片缩略图即可打开共享查看器，无需先执行 OCR。队列附件保持只读，不再展示解析来源和设置入口；草稿仍可显式提取文字。",
            "**整理设置导航与笔记间距**：设置按分组排列，持久开关统一右对齐，侧栏底部文字布局一并修正。笔记列表、详情和评论方向栏的标题对齐，文字编辑工具栏与输入区域减少多余留白，保留画布间距。",
            "**补充本地 MCP 故障诊断**：反馈诊断可包含连接阶段、错误类别及关联摘要，帮助定位初始化和清理问题。诊断不记录令牌、服务地址、请求正文或原始错误内容。"
          ],
          "notices": [
            "首次启动升级到 schema 52，为任务事件增加执行统计索引；历史较多时建立索引可能需要更长时间。升级前请关闭应用并备份完整数据库及相邻 `notes/` 目录；旧客户端不能打开升级后的数据库，降级需恢复升级前备份。",
            "本轮只更新 Desktop，Agent 保持 **0.15.1**，无需更新 Agent 包。仅提供标准六平台产物，不包含新的 LoongArch 预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Optional frosted-glass appearance**: Enable the effect in Appearance to let chat content scroll behind the header and give execution-checklist panels a frosted background. It is off by default and your choice persists across restarts; the effect is internal to the app and does not reveal the operating-system desktop.",
            "**Device-sharing metadata entry**: Open Device Sharing from the application menu or manager and connect to a separately running ShareServer to register this device, browse device and publication catalogs, publish capability or knowledge metadata, and revoke entries associated with this device ID. This first stage targets a local unauthenticated service, and device IDs are not ownership authentication; it does not execute remote capabilities or search, read, or download knowledge content, and permission options are metadata declarations only."
          ],
          "fixes": [
            "**Keep New note visible**: New note now stays in the Magic Notes page header rather than the list toolbar. You can start a note while viewing details or to-dos, with existing unsaved-draft protection retained.",
            "**Improve local OpenCode tool connection recovery**: Explicit transient MCP transport failures receive at most one retry after successful cleanup, and names with uncertain timeout or cancellation outcomes are retired so late initialization cannot replace a live connection. Authentication, protocol, and unknown errors are not retried, and prompts or tool calls are not replayed; this is not a new remote Agent recovery capability.",
            "**Reduce repeated work while typing and switching conversations**: Unchanged message history and task strips avoid repeated renders during typing, with stable callbacks reused across cached conversation switches. Streaming, scroll positions, drafts, and asynchronous attachment ownership remain intact, while execution-statistics indexes reduce event scans during polling.",
            "**Preview draft and queued images directly**: Click an image thumbnail to open the shared viewer without running OCR first. Queued attachments remain read-only and no longer show parsing-source or settings entries; drafts retain explicit text extraction.",
            "**Organize settings and tighten note spacing**: Settings navigation is grouped, persistent switches align to the right, and sidebar footer text layout is corrected. Note-list, detail, and comment-direction headers align, with excess gaps removed around text editing controls while preserving canvas spacing.",
            "**Add local MCP failure diagnostics**: Feedback diagnostics can include connection phases, error categories, and correlation digests to help identify initialization and cleanup failures. They do not record tokens, service addresses, request bodies, or raw error content."
          ],
          "notices": [
            "First launch upgrades to schema 52 and adds task-event indexes for execution statistics; building indexes may take longer with large histories. Close the app and back up the complete database and adjacent `notes/` directory first; older clients cannot open the upgraded database, so downgrading requires a pre-upgrade backup.",
            "This is a Desktop-only update; Agent remains **0.15.1**, with no Agent package update required. Only the six standard targets are released, with no new LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.15.5",
      "releasedAt": "2026-09-30",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**用标签整理魔法笔记**：可以手动给笔记添加标签，组合多个标签筛选时只显示同时匹配的笔记，并通过标签管理器重命名、合并或删除标签。笔记工具也支持创建、更新标签及按标签查找，标签变更仍需用户提出要求，不会自动分类。",
            "**打开笔记时保留列表和阅读空间**：当前笔记旁保留可调宽、可收起的笔记列表，详情使用紧凑的固定工具栏，切换笔记保留阅读位置并保护未保存草稿。“返回概览”回到全宽笔记卡片网格，窄窗口在列表与详情间切换。",
            "**智能心跳按变化回顾并提供有依据的建议**：心跳计划现在触发共享的增量回顾，不再另外生成一份心跳报告；可选择只更新记忆图谱，或根据本次回顾中的未决事项、分歧、修订和候选约定生成监督建议。建议支持核对来源、接受或忽略，接受未决事项会创建暂停任务，确认候选约定才会写为长期背景；建议生成失败可单独重试，不撤销已完成的回顾。",
            "**手动回顾默认只处理新增与变化内容**：点击“回顾”会沿用处理进度，无变化时不再调用模型；需要重读所选区间时，可从“更多”选择“重新整理…”并确认。重新整理不会推进或重置共享进度，自动回顾也不会续跑用户主动暂停的运行。"
          ],
          "fixes": [
            "**从快速笔记打开记录时可靠定位**：进入完整工作区后，等待目标记录挂载再滚动到对应内容。避免加载较慢时页面已打开却没有定位到所选记录。",
            "**再次在终端中打开 OpenCode 更快**：插件依赖改为在用户数据目录中缓存复用，不再每次启动都复制完整依赖树；随包版本变化后会更新缓存。首次准备和 OpenCode 自身启动仍需时间，模型配置、Skills 和 Ask 模式插件仍按每次启动生成。",
            "**减少 OpenCode 工具名称过长导致的失败**：缩短临时 MCP 名称及转发的自定义工具名称，并在同一会话中尽量保持稳定。并行请求仍使用独立占用的名称，不改变工具权限或执行范围。",
            "**滚动文件列表时保留工作区操作栏**：文件列表独立滚动，视图切换、刷新、新建和目录导航保持可见。没有 Git 的项目也保留“文件”入口。",
            "**取消后不再把上一轮问题并入新请求**：模型可见历史保留中断回复的已有文字，并明确标记取消或失败，避免用户问题成为没有回应的孤立输入。本地 OpenCode 复用会话时保留取消轮次，在下一轮附带一次中断说明，失败请求仍按原规则清理。",
            "**展开执行清单不再挤动聊天内容**：OpenCode 与 Continue 的清单详情改在锚定浮层中展开，长清单可在浮层内滚动。点击外部可关闭，按 Escape 关闭后焦点回到入口，进度保存规则不变。",
            "**笔记保存成功后退出编辑**：文字和画布记录保存修改后回到只读内容，自动评论完成后不会重新打开编辑器。保存失败时保留草稿，便于修改和重试。"
          ],
          "notices": [
            "首次启动升级到 schema 51，增加笔记标签存储，并将旧心跳报告中仍待确认的记忆迁为待处理约定，不自动确认或删除原记录。升级前请关闭应用并备份完整数据库及相邻 `notes/` 目录；旧客户端不能打开升级后的库，回退需恢复完整升级前备份。",
            "本轮只更新 Desktop，Agent 保持 **0.15.1**，无需重新下载 Agent 包。心跳回顾及有候选时的建议生成可能产生模型费用；项目与全局仍分别维护处理进度，会话级实时主动介入尚未实现。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Organize Magic Notes with tags**: Add tags manually, combine filters to find notes matching every selected tag, and rename, merge, or delete tags in the manager. Note tools also support creating, updating, and filtering tags when requested by the user; notes are not tagged automatically.",
            "**Keep the note list beside your reading area**: Open notes retain a resizable, collapsible list beside compact details with a sticky toolbar; switching notes preserves reading positions and protects unsaved drafts. Back to overview restores the full-width card grid, while narrow windows switch between list and detail.",
            "**Smart heartbeat reviews changes and offers evidence-backed suggestions**: Plans now trigger the shared incremental review instead of generating a separate heartbeat report; choose between updating memory and graphs only or generating suggestions from open items, disagreements, revisions, and candidate conventions. Inspect sources, accept, or dismiss suggestions; accepting an open item creates a paused task, and only confirming a convention saves it as long-term background. Suggestion generation can be retried independently without undoing a completed review.",
            "**Manual review defaults to new and changed content**: Review resumes from saved processing progress and makes no model calls when nothing has changed; choose Reanalyze from More and confirm to reread the selected period. Reanalysis neither advances nor resets shared progress, and automatic reviews no longer resume runs you explicitly paused."
          ],
          "fixes": [
            "**Reliably locate records opened from Quick Notes**: The full workspace now waits until the target record is mounted before scrolling to it. This avoids opening the page without locating the selected record when rendering is delayed.",
            "**Reopen OpenCode in the terminal faster**: Plugin dependencies are cached under user data instead of copying the entire tree on every launch, and refreshed when the bundled version changes. Initial preparation and OpenCode startup still take time; model settings, Skills, and Ask-mode plugins remain launch-specific.",
            "**Reduce failures from long OpenCode tool names**: Temporary MCP names and forwarded custom tool names are shorter and remain stable within a conversation where possible. Parallel requests retain separately leased names, with no changes to tool permissions or execution scope.",
            "**Keep workspace controls visible while scrolling**: The file list scrolls independently while view selection, refresh, creation actions, and directory navigation stay visible. Projects without Git retain the Files entry.",
            "**Keep cancelled questions out of the next request**: Model-visible history retains partial replies with explicit interruption or failure markers, rather than leaving their user messages unanswered. Reused local OpenCode sessions retain cancelled turns and receive a one-time interruption notice on the next turn; failed requests retain their existing cleanup behavior.",
            "**Expand execution checklists without shifting chat content**: OpenCode and Continue checklist details now open in an anchored floating panel with internal scrolling for long lists. Click outside to close, or press Escape to close and return focus to the trigger; saved progress behavior is unchanged.",
            "**Leave note editing after a successful save**: Saved text and canvas records return to read-only content, and automatic comments no longer reopen the editor. Failed saves retain your draft for correction and retry."
          ],
          "notices": [
            "First launch upgrades to schema 51, adds note-tag storage, and carries still-proposed memories from older heartbeat reports into pending conventions without automatically confirming or deleting the original records. Close the app and back up the complete database and adjacent `notes/` directory first; older clients cannot open the upgraded database, so rollback requires the complete pre-upgrade backup.",
            "This is a Desktop-only update; Agent remains **0.15.1**, with no Agent package download required. Heartbeat reviews and suggestion generation when candidates exist may incur model charges; project and global processing progress remain separate, and real-time conversation-level proactive intervention is not implemented."
          ]
        }
      }
    },
    {
      "version": "0.15.4",
      "releasedAt": "2026-09-29",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**按会话、项目和全局默认选择 Runtime 与模型**：执行方式和模型分别按“会话 → 项目 → 全局”解析，可在输入框两栏菜单中单独选择，或恢复跟随项目默认，并查看实际来源。项目设置与通道入口共用选择规则，排队请求在派发时重新解析；连接删除或不兼容时按规则回退，缺少密钥时保留选择并明确报错。",
            "**分开查看系统任务 Token 用量**：运行记录新增“系统任务”页签，将监督者、自动监督、魔法笔记、知识库和后台委派等用量与会话统计分开。支持按来源查看、手动刷新，并在新用量记录写入后自动刷新。"
          ],
          "fixes": [
            "**简化会话 Runtime 选择按钮**：移除图标与 Runtime 名称之间的圆点，避免选择模型后出现多余标记。菜单仍显示“本会话单独选择”和“恢复为项目默认”，选择规则不变。",
            "**补齐部分后台用量记录**：监督回顾、图片重新生成、知识图谱抽取和远程向量请求现在接入对应的用量记录。统计依据实际保存的供应商用量，不能补算过去未记录的调用或供应商未返回的 Token 数据。",
            "**避免 Runtime 重复携带指令与历史**：OpenCode 将受信任指令放入当前轮系统信息，复用 DeepSeek Harness 会话时不再重复注入已保留历史。Continue 识别已有历史封装并避免再次嵌套，保存用户轮次时移除工作模式前缀；远程 Continue 需要配套 Agent 更新。",
            "**减少失败请求对后续 OpenCode 对话的影响**：普通 OpenCode 请求失败且尚未产生回复记录时，会尝试清理该轮未回答的输入，避免下次复用会话时再次合并。已有回复、工具记录和其他并行会话保留。"
          ],
          "notices": [
            "首次启动会将数据库升级到 schema 49，迁移项目与会话的 Runtime/模型选择。升级前请关闭应用并备份完整用户数据，包括数据库和相邻 `notes/` 目录；旧客户端不能打开升级后的数据库，回退需恢复升级前备份。",
            "远程用户请先升级 Desktop 至 **0.15.4**，再下载并更新 Host 上的 Agent **0.15.1**。Agent 独立发布，Node、OpenCode 和 Continue 的锁定版本不变；已有进程不会热更新。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Choose Runtime and model through conversation, project, and global defaults**: Runtime and model resolve independently in that order, with a two-column composer menu for overrides, restoring project defaults, and viewing the actual source. Project and channel settings share the rules, and queued requests resolve again at dispatch; deleted or incompatible connections fall back according to those rules, while missing credentials retain the selection and produce an explicit error.",
            "**Separate system-task Token usage**: Run history adds a System tasks tab that separates Supervisor, automatic supervision, Magic Notes, knowledge, and background delegation usage from conversation totals. Browse by source, refresh manually, or receive an automatic refresh when new usage is recorded."
          ],
          "fixes": [
            "**Simplify the conversation Runtime button**: Remove the dot between the icon and Runtime name after choosing a model. The menu still identifies conversation overrides and offers Restore project defaults; selection rules are unchanged.",
            "**Capture additional background usage**: Supervisor reviews, image regeneration, knowledge-graph extraction, and remote embedding requests now feed their corresponding usage records. Statistics use stored provider usage and cannot reconstruct previously unrecorded calls or Token data omitted by a provider.",
            "**Avoid duplicate Runtime instructions and history**: OpenCode receives trusted instructions through the current turn's system information, and reused DeepSeek Harness sessions no longer receive history they already retain. Continue recognizes existing history envelopes instead of nesting them again and removes work-mode prefixes from stored user turns; remote Continue requires the matching Agent update.",
            "**Reduce carryover from failed OpenCode requests**: When a normal OpenCode request fails before producing a reply record, GoodBuddy attempts to remove its unanswered input so the next request does not merge it again. Existing replies, tool records, and other parallel conversations remain intact."
          ],
          "notices": [
            "First launch upgrades the database to schema 49 and migrates project and conversation Runtime/model selections. Close the app and back up complete user data, including the database and adjacent `notes/` directory, before upgrading; older clients cannot open the upgraded database, so rollback requires restoring a pre-upgrade backup.",
            "Remote users should upgrade Desktop to **0.15.4** before downloading and updating Hosts to Agent **0.15.1**. Agent releases remain separate, locked Node/OpenCode/Continue versions are unchanged, and existing processes are not hot-updated."
          ]
        }
      }
    },
    {
      "version": "0.15.3",
      "releasedAt": "2026-09-28",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**恢复直连模型手动压缩**：本地直连文本模型现在可使用“压缩上下文”，即使自动压缩已关闭或尚未达到触发阈值。摘要随会话保存并用于后续对话，同时保留最近轮次和完整历史；图像模型及远程直连入口不在本次范围内。",
            "**长请求不再丢失工具授权**：修复请求仍在执行时，知识、笔记、浏览器、配置及自定义 MCP 等工具在十分钟后失去访问权限的问题。授权随当前请求保持有效，取消或结束后仍会撤销，不改变 Ask 和 Execute 的权限边界。",
            "**区分取消和失败，避免重复提示**：请求取消或失败时保留已有回复，并只显示一次对应状态；取消采用中性提示，不再显示为错误。历史恢复保留终止状态，只有能恢复输入时才提供“重新编辑”。",
            "**减少聊天与笔记启动的重复处理**：置顶会话不再重复扫描附件，未变化的消息和附件引用不再重复写入，笔记启动复用正文检查结果。流式回复保留实时 Markdown、公式与代码显示，不以延迟富文本换取速度，也不宣称整体渲染耗时已显著下降。",
            "**侧栏搜索与快捷键提示更清楚**：“新建对话”和可展开搜索共用一行，应用与设置入口、分隔线和滚动条保持一致布局。输入框提示只显示已成功注册的快捷键，设置保存后及时更新，较晚返回的启动信息不会恢复已关闭的提示。",
            "**笔记和工作栏操作更紧凑**：笔记采集的按钮靠右排列，范围说明按需查看，输入校验贴近对应字段；窄窗口中的保存按钮不再拉伸。文件差异操作与路径分组，浏览器前往按钮使用图标并保留提示和可访问名称。",
            "**精简桌面安装包内容**：已打入界面产物的依赖不再重复作为运行依赖分发，DS Web 资源排除类型声明、调试映射和调试符号。保留运行资源与许可证，减少重复内容和散落文件，具体安装包大小以各平台发布资产为准。"
          ],
          "notices": [
            "本次仅更新 Desktop，继续使用已发布的 Agent **0.15.0**，无需为本版重新安装 Agent。",
            "手动压缩会调用所选摘要模型，可能产生费用。数据库仍为 schema 48；从旧版升级前请备份完整数据库和相邻 `notes/` 目录。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Restore manual compaction for direct models**: Local direct text-model conversations can now use Compact context even when automatic compaction is disabled or its threshold has not been reached. Summaries are saved with the conversation and reused in subsequent requests while recent turns and full history remain available; image models and remote direct-model entry points are outside this change.",
            "**Keep tool access during long requests**: Fixes active requests losing access to knowledge, notes, browser, configuration, and custom MCP tools after ten minutes. Authorization now lasts for the current request and is still revoked on cancellation or completion, without changing Ask or Execute permission boundaries.",
            "**Distinguish cancellation from failure without duplicate notices**: Cancelled and failed requests retain existing replies and show a single terminal status; cancellation uses neutral feedback rather than an error. Restored history preserves that status, and Edit again is offered only when the input can be recovered.",
            "**Avoid redundant chat and note startup work**: Pinning conversations no longer rescans attachments, unchanged messages and attachment references avoid repeat writes, and note startup reuses body inspection results. Streaming replies retain live Markdown, math, and code rendering without delaying rich text or claiming a proven overall rendering speedup.",
            "**Clearer sidebar search and shortcut hints**: New conversation and expandable search share one row, with consistent Apps and Settings entries, dividers, and scrollbars. Composer hints show only successfully registered shortcuts, reflect saved settings, and cannot be restored by a late startup response after being disabled.",
            "**More compact note and workbar actions**: Note capture uses right-aligned actions, on-demand scope help, and field-specific validation; Save buttons no longer stretch in narrow windows. File-diff actions are grouped with their path, and the browser Go button uses an icon while retaining its tooltip and accessible name.",
            "**Trim desktop package contents**: Dependencies already bundled into the interface are no longer duplicated as runtime dependencies, and DS Web resources exclude type declarations, source maps, and debug symbols. Runtime assets and licenses remain included, reducing duplicate content and loose files; final sizes are listed with each platform's release assets."
          ],
          "notices": [
            "This is a Desktop-only update and continues to use published Agent **0.15.0**. Reinstalling the Agent is not required for this release.",
            "Manual compaction calls the selected summary model and may incur charges. The database remains on schema 48; back up the complete database and adjacent `notes/` directory before upgrading from older versions."
          ]
        }
      }
    },
    {
      "version": "0.15.2",
      "releasedAt": "2026-09-28",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**让 Agent 查阅故事图谱**：启用监督者并分配 Story Graph 工具后，可让模型查找已保存的决定、变化过程和来源片段，在继续工作前核对已有依据。支持直连模型、本地 DeepSeek Harness，以及本地和远程 OpenCode、Continue；读取不修改图谱，也不代表完整历史重建或跨范围事实合并。"
          ],
          "fixes": [
            "**继续讨论前编辑完整消息**：从监督结果继续讨论时，可查看并修改即将发送的完整上下文和问题，并确认目标会话。发送以编辑后的可见内容为准，不会重新附加已删除的上下文。",
            "**简化笔记采集界面**：采集和快速追加面板移除重复标签、范围提示及重复入口，保留内容编辑、目标选择和必要的校验错误。输入框仍有可访问名称，保存行为和来源关联不变。",
            "**保留窗口异常退出线索**：诊断日志现在记录渲染进程退出原因和退出码，便于排查窗口异常。此项改善诊断信息，不代表已修复所有导致窗口退出的问题。",
            "**下载与入门信息更容易查找**：双语网站调整产品介绍、使用场景和设置问答，下载入口紧接首页介绍。移动端展示同步优化，DeepSeek Harness 与 LoongArch 仍保留预览标识。"
          ],
          "notices": [
            "远程 Story Graph 需要配套更新 Desktop **0.15.2** 和 Agent **0.15.0**，先升级桌面端，再更新 Host。",
            "数据库保持 schema 48；从更早版本升级前仍应备份完整用户数据，包括数据库和相邻 `notes/` 目录。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Let agents consult Story Graph**: Enable Supervisor and assign the Story Graph tools to let agents look up saved decisions, their evolution, and source fragments before continuing work. Supports direct models, local DeepSeek Harness, and local or remote OpenCode and Continue; reads do not modify the graph or provide complete historical reconstruction or cross-scope fact merging."
          ],
          "fixes": [
            "**Edit the complete follow-up message**: Continuing from Supervisor results now shows the full context and question for editing, together with the destination conversation. Sending uses the visible draft without reattaching context you removed.",
            "**Simpler note capture controls**: Capture and quick-append panels remove repeated labels, scope hints, and duplicate entry points while retaining editing, destination selection, and actionable validation errors. Inputs keep accessible names, and saving and source links behave as before.",
            "**Retain window-exit diagnostics**: Diagnostic logs now retain renderer exit reasons and exit codes to help investigate unexpected window exits. This improves diagnostics rather than claiming to fix every underlying cause.",
            "**Clearer downloads and setup guidance**: The bilingual website refreshes product explanations, usage scenarios, and setup FAQs, with downloads immediately after the introduction. Mobile layouts improve while DeepSeek Harness and LoongArch retain their preview labels."
          ],
          "notices": [
            "Remote Story Graph requires Desktop **0.15.2** and Agent **0.15.0**. Upgrade Desktop before updating Hosts.",
            "The database remains on schema 48. Before upgrading from older versions, back up complete user data, including the database and adjacent `notes/` directory."
          ]
        }
      }
    },
    {
      "version": "0.15.1",
      "releasedAt": "2026-09-28",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**按需关闭桌面通知**：在“平台功能 / 通用设置”中可随时开关系统任务提醒，默认保持开启且保存后立即生效。关闭后仍保留应用内提示，通用设置也按快捷键、会话与通知、工作目录与下载分组，便于查找。",
            "**连接本地 Obsidian 仓库**：在 MCP 设置中测试并启用全部已注册仓库或指定文件夹，即可在受支持的本地对话中搜索、阅读笔记，并在 Execute 中写回结果，无需安装 Obsidian 插件。默认关闭，支持直连模型及 GoodBuddy 管理的本地 OpenCode、Continue，暂不支持 DeepSeek Harness 和远程执行。",
            "**把对话保存到魔法笔记**：单条 AI 回答或完整会话可先预览、编辑，再追加到已有笔记或创建新笔记，保存后可返回来源会话。右侧工作栏新增笔记入口，支持搜索与快速追加，切换会话时保留未提交草稿。",
            "**从输入区打开本地 Runtime 客户端（预览）**：本地项目选择 Continue 或 OpenCode 后可打开工作栏原生终端，选择 DS 后可在系统浏览器打开官方 Web 客户端，使用当前项目和模型。客户端是独立会话，不复制 GoodBuddy 聊天历史；DS 网页关闭后服务仍运行，可在 GoodBuddy 中停止，远程快捷入口尚不支持。"
          ],
          "fixes": [
            "**流式回复时会话排序稳定**：对话按最近一条消息的创建时间排序，不再因流式文本或工具进度更新反复交换位置。新消息仍会推动会话前移，置顶会话继续排在前面。",
            "**从监督结果直接继续讨论**：监督反馈和图谱来源可直接发起后续讨论，编辑问题后将相关回顾结论与来源正文带入目标会话，无需先打开来源。界面移除重复上下文说明和知识实体写回控件，自动监督没有活动时不再显示空指标。",
            "**知识库列表与操作更清楚**：内容来源和文档索引合并展示，单文档来源不再重复占行，目录分组、失败来源和统一搜索仍可使用。整理重新处理、移除与任务历史入口，减少重复菜单和不明确的操作。",
            "**监督回顾避免重复运行**：手动开始、自动触发和继续回顾共用运行状态，已有回顾执行时会定位当前任务，不再叠加新回顾。可从活动记录取消，已保存批次保留；故事线事件、实体和关系改为独立页签浏览。",
            "**修复弹窗与窄窗口操作**：恢复删除确认按钮的可见样式，修正弹层遮挡、控件状态及窄布局滚动。助手工作栏的添加按钮不再被滚动箭头覆盖，设置和知识库中的操作保持可达。",
            "**本地 Runtime 启动更可靠**：修复便携包缺少 DS Web 入口及依赖导致无法打开浏览器的问题。Continue 关闭自动更新后不再停留在更新检查提示，DS 输入区保留停止服务操作而不再常驻重复说明。",
            "**补齐 Runtime 遥测关闭设置**：DS Web 与配套 Agent 的远程 Runtime 启动统一应用遥测关闭配置。模型、MCP 和用户主动联网请求保持可用，已有自动更新关闭设置不变。"
          ],
          "notices": [
            "首次启动会将数据库升级至 schema 48。升级前请备份完整用户数据，包括数据库及相邻 `notes/` 目录；旧客户端不能打开升级后的数据库，回退需使用升级前备份。",
            "远程用户请先升级 Desktop 至 **0.15.1**，再下载并更新 Host 上的 Agent **0.15.0**。已有进程不会热更新，Agent 独立发布，不替代桌面端 Latest。",
            "原生客户端仍为预览，其 Skills/MCP 与 GoodBuddy 会话能力并非完全等价；Continue 和 DS Web 首次启动可能需要准备标准 Node 环境。真实模型请求可能产生费用，其他平台与安装包验收以本次发布流程结果为准。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Control desktop notifications**: Turn system task notifications on or off under Platform Features / General; the default remains on and saved changes take effect immediately. In-app notifications remain available when system notifications are off, and General settings are grouped into shortcuts, conversations and notifications, and workspace and downloads.",
            "**Connect local Obsidian vaults**: Test and enable all registered vaults or a selected folder in MCP settings to search and read notes in supported local conversations, and write results back in Execute, without an Obsidian plugin. Disabled by default, this supports direct models and GoodBuddy-managed local OpenCode and Continue, but not DeepSeek Harness or remote execution.",
            "**Save conversations to Magic Notes**: Preview and edit an AI response or a complete conversation before appending it to an existing note or creating a new one, with links back to the source conversation. The new Notes workbar entry supports search and quick appends while preserving unsent drafts across conversation switches.",
            "**Open local Runtime clients from the composer (preview)**: In a local project, open a native workbar terminal for Continue or OpenCode, or the official DS Web client in your system browser, using the current project and model. These are independent sessions without copied GoodBuddy chat history; closing the DS page leaves its service running until stopped in GoodBuddy, and remote shortcuts are not supported yet."
          ],
          "fixes": [
            "**Stable conversation order while replies stream**: Conversations are ordered by the creation time of their latest message rather than repeatedly swapping positions as text or tool progress arrives. New messages still move conversations forward, with pinned conversations first.",
            "**Continue directly from Supervisor results**: Start a follow-up from supervision feedback or graph sources, edit the question, and send relevant review conclusions and source text to the target conversation without opening the source first. Redundant context explanations and knowledge-entity writeback controls are removed, and Automatic supervision hides empty metrics until activity exists.",
            "**Clearer knowledge lists and actions**: Sources and indexed documents now share one list, without duplicate rows for single-document sources, while retaining directory groups, failed sources, and unified search. Reprocessing, removal, and task-history actions are organized to reduce duplicate menus and ambiguous operations.",
            "**Avoid overlapping Supervisor reviews**: Manual starts, automatic triggers, and resumed reviews share the active-run state, directing you to existing work instead of starting an overlapping review. Cancel from Activity while retaining saved batches, and browse story-graph events, entities, and relations in separate tabs.",
            "**Usable dialogs and narrow layouts**: Restores visible destructive-confirmation buttons and fixes overlay occlusion, control states, and scrolling in narrow layouts. Workbar scroll arrows no longer cover the add button, and settings and knowledge actions remain reachable.",
            "**More reliable local Runtime startup**: Fixes missing DS Web entry files and dependencies in portable packages that prevented browser launch. Continue no longer remains on an update-check message when automatic updates are disabled, and the DS composer keeps its stop action without persistent duplicate help text.",
            "**Consistent Runtime telemetry opt-out**: DS Web and remote Runtimes in the matching Agent apply the shared telemetry opt-out settings at startup. Model, MCP, and user-requested network access remain available, and existing automatic-update opt-outs are unchanged."
          ],
          "notices": [
            "First launch upgrades the database to schema 48. Back up complete user data, including the database and adjacent `notes/` directory, before upgrading; older clients cannot open the upgraded database, so rollback requires a pre-upgrade backup.",
            "Remote users should upgrade Desktop to **0.15.1** before downloading and updating Hosts to Agent **0.15.0**. Existing processes are not hot-updated, and the independent Agent release does not replace Desktop Latest.",
            "Native clients remain a preview and do not offer full parity with GoodBuddy conversation Skills/MCP capabilities; Continue and DS Web may need to prepare a standard Node environment on first launch. Real model requests may incur charges, and other-platform and packaged-app acceptance depend on this release's validation results."
          ]
        }
      }
    },
    {
      "version": "0.13.16",
      "releasedAt": "2026-09-26",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**避免重复整理历史**：普通数据库结构升级不再因为存在空闲页而重新整理历史或执行全库空间回收。启动页会区分结构更新、旧内容转换和实际空间回收，真正需要转换的旧数据仍按原流程处理并保留已有内容。",
            "**监督回顾持续处理**：回顾较长历史时，不再每隔固定时长自动暂停，减少反复点击继续的操作。仍可主动暂停并从已保存批次继续，单次模型请求的超时和并发设置保持有效。",
            "**监督页面布局更清楚**：修复工作回顾空态挤在控件旁的问题，运行状态单独显示，减少顶部重复说明。活动记录突出处理阶段、已保存批次和剩余来源，次要统计可展开查看。"
          ],
          "notices": [
            "本次仅更新 Desktop，Agent 保持 **0.13.4**，无需为这些修复更新 Host 环境。从 Desktop **0.13.15** 升级不新增数据库格式迁移；从更早版本升级时，仍应备份完整用户数据，包括数据库和相邻 `notes/` 目录。",
            "监督回顾只读且不调用工具，但会使用已配置模型并可能产生费用。取消固定时长自动暂停后，较长回顾可能连续执行更多批次；可主动暂停，模型结果仍需结合保存的来源核对。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Avoid repeated history cleanup**: Ordinary database structure upgrades no longer reorganize history or reclaim the entire database merely because free pages exist. The startup page distinguishes structure updates, legacy data conversion, and actual space reclamation, while older data that genuinely needs conversion still follows the existing process without losing its contents.",
            "**Keep Supervisor reviews running**: Reviews of longer histories no longer pause automatically at fixed time intervals, reducing repeated manual continuation. You can still pause and resume from saved batches, and per-request model timeouts and concurrency settings remain effective.",
            "**Clearer Supervisor layouts**: Fixes the Work review empty state appearing beside its controls, gives run status its own area, and removes repeated introductory text. Activity emphasizes processing stages, saved batches, and remaining sources, with secondary statistics available in expandable details."
          ],
          "notices": [
            "This release updates Desktop only. Agent remains at **0.13.4**, with no Host update required for these fixes; upgrading from Desktop **0.13.15** adds no database format migration. When upgrading from older versions, back up complete user data, including the database and adjacent `notes/` directory.",
            "Supervisor reviews are read-only and cannot call tools, but use the configured model and may incur charges. Without fixed-duration automatic pauses, longer reviews may process more batches continuously; you can pause manually, and results should still be checked against their saved sources."
          ]
        }
      }
    },
    {
      "version": "0.13.15",
      "releasedAt": "2026-09-26",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**待办列表调宽与分组折叠**：在魔法笔记的待办页面，可拖动列表与详情之间的分隔条，调整后的宽度会保留。可按来源笔记独立折叠待办分组，当前选中任务的详情不会因此清空。",
            "**监督回顾进度与续跑**：回顾较长历史时，可查看处理阶段和已保存批次，并暂停后从活动记录继续。监督者设置提供并发、超时、批次大小、执行预算和响应容量，便于按模型能力调整。"
          ],
          "fixes": [
            "**回顾覆盖与结果保存**：监督回顾改为分页读取所选历史，分批保存事实与来源，避免只处理截断后的输入。修复旧实体引用、汇总结果校验和长摘要报错，已保存的成功批次可在继续时复用。",
            "**监督者页面更清楚**：工作回顾、自动监督、活动记录和设置各自承载对应内容，计划通过统一弹窗创建和编辑。工作回顾与活动记录使用面板可用全宽，侧栏监督反馈位于项目概览下方。",
            "**直连模型搜索更完整**：文件搜索支持 ripgrep 原生参数，保留原生输出与退出码，大结果可分页续读。Ask 保持工作区只读，Execute 使用当前账号权限，无需另行安装 ripgrep。",
            "**远程结果未知后继续对话**：远程请求结果未知时，再次显式发送消息会创建新的会话绑定并带入已有历史，不再复用失效绑定。自动恢复仍只附加原请求，不会自动重发结果未知的任务。",
            "**布局与队列操作修正**：项目选择列表可浮出侧栏，窄窗口下自动收窄；助手工作栏恢复完整拖动命中区域，笔记缩略图与内容边缘对齐。没有附件的排队消息不再显示“查看附件”。",
            "**远程任务完成与更新清理**：配套更新 Agent 0.13.4 后，远程任务等待模型交付确认再结束，空闲 Runtime 回收失败不再把已完成回答改成结果未知。更新 Host 环境时清理已确认停用的旧 Agent 程序和所属孤儿进程，保留会话历史、活动安装及状态不明的安装。"
          ],
          "notices": [
            "首次启动会将数据库升级至 schema 47。升级前请备份完整用户数据，包括数据库及相邻 `notes/` 目录；旧客户端不能打开升级后的数据库，回退需使用升级前备份。",
            "监督回顾只读且不调用工具，但会调用已配置模型并可能产生费用。分页覆盖不代表模型识别了每个事实，请结合保存的来源核对结果。",
            "远程用户请先升级 Desktop 至 **0.13.15**，再下载并更新 Host 上的 Agent **0.13.4**。已有旧进程不会热更新，仍在运行或状态不明的旧安装不会被强制删除；Agent 独立发布，不替代桌面端的 Latest 版本。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Resizable to-do list and collapsible groups**: In Magic Notes, drag the divider between the to-do list and details to adjust and retain its width. Collapse each source-note group independently without clearing the selected task's details.",
            "**Review progress and resume controls**: When reviewing longer histories, inspect execution stages and saved batches, pause, and continue from Activity. Supervisor settings expose concurrency, timeouts, batch sizes, execution budgets, and response capacity to match your model's capabilities."
          ],
          "fixes": [
            "**Review coverage and saved results**: Supervisor now pages through the selected history and saves facts and sources in batches instead of processing truncated input. Fixes address historical entity references, summary validation, and long-summary errors; successfully saved batches can be reused when continuing.",
            "**Clearer Supervisor pages**: Work review, Automatic supervision, Activity, and Settings each contain their corresponding controls and results, with a shared dialog for creating and editing plans. Work review and Activity use the available panel width, and sidebar supervision feedback appears below the project overview.",
            "**More complete direct-model searches**: File searches support native ripgrep arguments, preserve native output and exit codes, and provide paged access to large results. Ask remains read-only within the workspace, while Execute uses current-account permissions; no separate ripgrep installation is needed.",
            "**Continue remote conversations after an uncertain result**: Sending another message explicitly after an uncertain remote outcome creates a new session binding with existing history instead of reusing the failed binding. Automatic recovery still attaches only to the original request and never automatically resends a task with an uncertain outcome.",
            "**Layout and queue controls**: The project picker can extend beyond the sidebar and narrows to fit small windows; the assistant workbar restores its full resize hit area, and note thumbnails align with content edges. Queued messages without attachments no longer show “View attachments.”",
            "**Remote task completion and update cleanup**: With Agent 0.13.4, remote tasks wait for model delivery acknowledgement before completing, and idle Runtime cleanup failures no longer turn completed answers into uncertain outcomes. Host environment updates remove verified inactive Agent payloads and their owned orphan processes while retaining conversation history, active installations, and installations with uncertain state."
          ],
          "notices": [
            "The first launch upgrades the database to schema 47. Back up complete user data, including the database and adjacent `notes/` directory, before upgrading; older clients cannot open the upgraded database, and rollback requires a pre-upgrade backup.",
            "Supervision reviews are read-only and cannot call tools, but use the configured model and may incur charges. Paged coverage does not guarantee that the model identifies every fact; check results against their saved sources.",
            "Remote users should upgrade Desktop to **0.13.15** before downloading and updating Host environments to Agent **0.13.4**. Existing older processes are not hot-updated, and active or uncertain older installations are not forcibly removed; Agent releases remain separate and do not replace the Desktop release marked Latest."
          ]
        }
      }
    },
    {
      "version": "0.13.13",
      "releasedAt": "2026-09-23",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**减少流式活动历史的重复读取**：活动历史较多时，连续保存流式更新不再每次重新读取整份历史。保留已有记录与顺序，保存失败后仍可重试。",
            "**修复监督反馈侧栏布局**：固定目标与刷新操作使用标题旁的紧凑按钮，目标显示会话或任务名称。长标题和窄侧栏下保持内容可读，并保留键盘焦点提示。"
          ],
          "notices": [
            "本次仅更新 Desktop，Agent 保持 **0.13.3**，无需因此更新 Host 环境。从 Desktop 0.13.12 升级不新增数据库格式迁移。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Reduce repeated activity-history reads during streaming**: With a large activity history, consecutive streaming saves no longer reread the entire history on every update. Existing records and ordering are preserved, and failed saves remain retryable.",
            "**Fix the supervision feedback sidebar layout**: Pin and refresh actions now use compact buttons beside the heading, and targets show conversation or task names. Long titles remain readable in narrow sidebars, with visible keyboard focus."
          ],
          "notices": [
            "This release updates Desktop only. Agent remains at **0.13.3**, with no Host update required for these fixes; upgrading from Desktop 0.13.12 adds no database format migration."
          ]
        }
      }
    },
    {
      "version": "0.13.12",
      "releasedAt": "2026-09-23",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**监督者工作回顾与故事线图谱**：从应用中心启用监督者后，可按范围和时间回顾工作，查看历史结果及其事件、实体、关系和来源。会话侧栏可固定到会话或任务并定位对应图谱，核对后可继续讨论或预览写入本地知识库实体。",
            "**自动监督与活动记录**：在设置中保存并启用每日或每周计划，自动回顾窗口内新增、修改或尚未处理的来源片段；没有待处理输入时不调用模型，也不生成重复报告。活动页显示手动及自动运行、各阶段状态和失败原因，并可打开对应历史结果；手动回顾会重新整理所选区间。",
            "**Mermaid 完整图片导出**：在大图窗口中可下载完整 PNG，导出不受当前缩放和拖动位置影响。放大后也可滚动查看图表各个边缘。"
          ],
          "fixes": [
            "**远程任务结束后释放空闲进程**：更新配套 Agent 后，任务完成会回收空闲 Runtime，下一轮恢复对话历史；共享进程仍有任务或待答问题时继续保留。被更新替换的 Agent 等已有任务完成后退出，当前 Agent 保持常驻；远程项目移除技术预览标识，仍默认关闭。",
            "**用量与回复时长更容易查看**：项目和会话用量先显示汇总，展开后查看 Runtime 与模型明细。切回已查看的会话时立即显示缓存的累计回复时长，并在后台刷新。",
            "**减少流式回复期间的磁盘写入**：活动更新只保存变化的记录，避免反复重写整份历史。升级保留已有活动内容和顺序。",
            "**工作区和笔记操作更清楚**：Git 分支选择器悬浮显示，不再挤动文件列表和提交历史。笔记索引开关移到返回按钮旁，索引与 AI 评论开关都显示文字。",
            "**通知更简洁**：成功与提示通知采用紧凑布局，两秒后自动消失。错误通知继续保留，直到手动关闭。",
            "**浏览器关闭重试后清除旧错误**：关闭页签失败时仍保留页签，方便重试。重试成功后清除原错误提示。"
          ],
          "notices": [
            "监督者默认关闭，缺失启用设置时也按关闭处理，已明确保存的选择保留；升级后请在应用中心检查开关。关闭后保留计划和历史，停止接受新回顾，已在执行的运行可完成；启用应用不会自动创建计划。",
            "监督回顾只读且不调用工具，但生成内容会调用已配置模型并可能产生费用。当前不支持分钟间隔、用户取消、逐事件完整回放、Experiment 目标或人工修改的时间序列审计；输入预算与回顾窗口仍有限制，删除报告不会重置自动进度。",
            "首次启动会升级数据库。升级前请备份完整用户数据，包括数据库和相邻 notes/ 目录；旧客户端不能打开升级后的数据库，回退需使用升级前备份。",
            "远程用户请先升级 Desktop **0.13.12**，再下载并更新 Host 上的 Agent **0.13.3**。已有旧进程不会热更新，Execute 继续使用所选 SSH 账号的权限。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Supervisor work reviews and story graphs**: Enable Supervisor in App Center to review a selected scope and time range, inspect past results, and trace their events, entities, relations, and sources. The conversation sidebar can pin a conversation or task and open its matching graph, with confirmed continuation and previews for writing entities to a local knowledge base.",
            "**Automatic supervision and activity history**: Save and enable a daily or weekly plan to review new, changed, or unprocessed source portions within its window; checks without pending input make no model call and create no duplicate report. Activity shows manual and automatic runs, stage status, failures, and links to the exact saved result; manual review reprocesses the selected period.",
            "**Complete Mermaid image export**: Download the entire diagram as PNG from the large-diagram viewer, independently of zoom and pan. All diagram edges remain reachable when zoomed in."
          ],
          "fixes": [
            "**Release idle processes after remote tasks**: With the matching Agent update, completed tasks release idle Runtimes and the next turn restores conversation history; shared processes remain alive while tasks or pending questions are active. Replaced Agents finish existing work before exiting while the current Agent stays resident; Remote Projects loses its technical preview label and remains disabled by default.",
            "**Clearer usage totals and reply duration**: Project and conversation usage shows totals first, with Runtime and model details available by expanding each row. Returning to a viewed conversation immediately restores its cached cumulative reply duration while refreshing it in the background.",
            "**Fewer disk writes during streaming replies**: Activity updates save only changed records, avoiding repeated writes of the entire history. Upgrades preserve existing activity content and order.",
            "**Clearer workspace and note controls**: The Git branch picker floats above the workspace without displacing files or commit history. The note index toggle sits beside Back, and both index and AI comment toggles display text labels.",
            "**Simpler notifications**: Success and informational notifications use a compact layout and dismiss after two seconds. Errors remain visible until manually dismissed.",
            "**Clear browser close errors after a successful retry**: A failed close keeps the tab available for retry. Once the retry succeeds, the earlier error is cleared."
          ],
          "notices": [
            "Supervisor defaults to off, including when the enabled preference is missing; explicit saved choices are preserved. Check its App Center switch after upgrading: disabling it retains plans and history and blocks new reviews while in-flight runs may finish; enabling it does not create plans automatically.",
            "Supervision reviews are read-only and cannot call tools, but generating content uses the configured model and may incur charges. Minute intervals, user cancellation, full event-by-event replay, Experiment targets, and a chronological audit of manual edits are not supported; input budgets and review windows still apply, and deleting reports does not reset automatic progress.",
            "The first launch upgrades the database. Back up complete user data, including the database and adjacent notes/ directory, before upgrading; older clients cannot open the upgraded database, and rollback requires a pre-upgrade backup.",
            "Remote users should upgrade Desktop to **0.13.12** before downloading and updating Host environments to Agent **0.13.3**. Existing older processes are not hot-updated, and Execute retains the selected SSH account's permissions."
          ]
        }
      }
    },
    {
      "version": "0.13.11",
      "releasedAt": "2026-09-22",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**优化发布资源上传**：发布资源上传支持 OSS 传输加速，改善跨地域网络环境下的大文件上传稳定性。"
          ],
          "notices": []
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Improve release asset uploads**: Release asset uploads now support OSS transfer acceleration for better stability when uploading large files across regions."
          ],
          "notices": []
        }
      }
    },
    {
      "version": "0.13.10",
      "releasedAt": "2026-09-21",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**任务统计使用后台只读读取**：执行统计通过独立只读 worker 查询 WAL 数据，避免较重的统计 SQL 占用 Electron 主进程；统计缓存有明确的容量、有效期和失效边界。",
            "**任务中心切换更顺畅**：隐藏任务面板保留运行状态并降低后台渲染优先级，切换任务中心、工作区、浏览器和成果页时减少重复渲染。"
          ],
          "fixes": [
            "**修复 Agent 回复期间任务中心阶段性卡顿**：减少任务事件写入、统计刷新和 tab 切换渲染叠加造成的 UI 阻塞。",
            "**完善运行时存储边界**：明确 WAL、主数据库连接、启动迁移 worker、统计 worker 和 Renderer 的职责，避免把启动迁移缓存误当作运行时统计缓存。",
            "**删除对话确认更简洁**：移除重复的底部警告文案，保留明确的取消和永久删除操作。"
          ],
          "notices": [
            "本次仅更新 Desktop，Agent 保持 **0.13.2**，无需更新 Host 环境。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Run execution statistics through a background read-only path**: Execution statistics now query WAL data through a dedicated read-only worker, keeping heavier aggregation SQL off the Electron Main thread with explicit cache capacity, lifetime, and invalidation rules.",
            "**Smoother Task Center switching**: Hidden task panels preserve their running state while receiving lower-priority background rendering, reducing repeated work when switching between Task Center, Workspace, Browser, and Results."
          ],
          "fixes": [
            "**Reduce intermittent Task Center stalls during Agent replies**: Reduced UI contention caused by task-event persistence, statistics refreshes, and tab-switch rendering occurring at the same time.",
            "**Clarify runtime storage boundaries**: Documented the separate responsibilities of WAL, the business database connection, the startup migration worker, the statistics worker, and Renderer scheduling.",
            "**Simplify conversation deletion confirmation**: Removed the duplicate bottom warning text while keeping clear Cancel and Permanently Delete actions."
          ],
          "notices": [
            "This release updates Desktop only. Agent remains at **0.13.2**; no Host update is required."
          ]
        }
      }
    },
    {
      "version": "0.13.9",
      "releasedAt": "2026-09-21",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**修复重启后重复显示历史迁移**：修复已有笔记的用户正常聊天后，再次打开应用仍进入迁移页面的问题。已完成升级的数据继续按新格式保存，正常读写产生的数据库空闲页不再触发迁移；真正的旧数据仍会正常升级。"
          ],
          "notices": [
            "本次仅更新 Desktop，Agent 保持 0.13.2，无需因此更新 Host 环境。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Stop repeated history migration after restart**: Fixed users with existing notes seeing the migration screen again after normal chat activity and reopening the app. Upgraded data continues to use the new storage format, and free database pages from normal writes no longer trigger migration; genuinely older data still upgrades normally."
          ],
          "notices": [
            "This release updates Desktop only. Agent remains at 0.13.2, with no Host environment update required for this fix."
          ]
        }
      }
    },
    {
      "version": "0.13.8",
      "releasedAt": "2026-09-20",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**设置页面滚动更顺手**：设置分类标题和当前分类的页签会固定在内容区顶部，平台功能、能力与工具、消息通道以及 Runtime 的子页签滚动长表单时仍可切换。低频配置继续放在默认收起的高级设置中。",
            "**复制模型连接**：在模型连接详情中复制已有连接，地址、模型、协议、请求定制和认证设置会带到新连接。已保存的 API Key 由主进程从加密凭据复制，不需要再次输入，也不会暴露给界面。",
            "**分页画布笔记**：在魔法笔记中混排正文、笔迹、高亮和图片，也可导入 PDF 页面进行批注。支持手动保存后继续编辑、25%–300% 缩放、适应宽度和 PDF 导出。",
            "**选择送给 AI 的画布页数**：在笔记设置中选择发送前 1～8 页，默认 1 页，文字与图片使用相同范围。保存和导出仍保留完整画布，最多支持 50 页；页面图像分析需要支持图片输入的模型。",
            "**接入 HTTP PaddleOCR-VL**：在文档解析设置中配置 OCR 服务，供聊天附件和知识库导入共用，无需在本机加载该服务的模型权重。可用实际文件测试解析结果，并配置服务支持的版面、表格等处理选项。",
            "**文档图文可以保留和复用**：附件原件及解析插图随引用保存，重开会话后可继续查看正文、图片和详情。可选择文档中的图片加入指定会话草稿，也可对图片显式提取文字；旧历史中已缺失的文件不会自动补齐。",
            "**统一管理内置应用**：从应用中心打开管理页面，调整四个内置应用的顺序，菜单和侧栏同步排列。魔法笔记与本机推理监控可设置启用和常驻，知识库与智能心跳始终显示；私有化应用市场仍属后续规划。",
            "**笔记与待办分开浏览**：笔记详情连续展示记录，通过缩略图索引定位，并可独立收起索引和 AI 评论栏。待办页提供搜索、状态筛选和详情，返回来源笔记后可恢复原任务位置。",
            "**对话中查看执行清单**：OpenCode 和 Continue 的原生清单显示在对话顶部，可展开查看进度，并随会话保存。清单由 Runtime 更新，不是手动待办；停止任务不会把未完成条目自动标为完成。",
            "**远程项目支持 Continue**：更新配套 Agent 后，可在托管 SSH 项目中选择 Continue，使用模型、执行清单和 Execute 模式下的会话 MCP 工具。支持取消、断线恢复以及重启后继续对话，Ask 保持只读。",
            "**常用会话置顶与完成提示**：会话可置顶并在重启后保留。后台会话成功完成后，在本次客户端运行期间保留提示，进入对应聊天后清除。",
            "**本机推理监控与任务统计**：可查看本地服务状态，并执行各服务支持的停止、重启或释放操作；向量服务显示独立进程 CPU 和内存，语音与 OCR 保留共享进程的统计限制。任务中心补充状态数量和筛选，便于找到正在运行或需要处理的任务。",
            "**说明按需查看**：页面和设置中的补充说明移到标题或字段旁的帮助入口，支持鼠标与键盘操作。减少常驻说明文字，同时保留关键操作后果和错误提示。"
          ],
          "fixes": [
            "**历史数据迁移只在需要时执行**：首次升级旧数据库时仍会整理重复的执行记录；数据库完成当前版本升级后，聊天、退出并重新打开应用不会再次显示历史迁移页面。",
            "**模型 URL 错误更容易处理**：保存模型连接时，空白、过长或不支持的地址会明确指出具体连接，并在 URL 字段旁保留可访问的错误提示，不再只显示笼统的“Invalid URL”。",
            "**较早会话和长历史不再被数量截断**：会话列表与搜索不再只覆盖最近 100 个会话，连续发送不再截断超过 500 条的历史。模型上下文限制仍然有效，已经丢失的历史无法恢复。",
            "**保留草稿和画布编辑**：恢复队列文字时不再覆盖新输入，画布删页保留正文与清单格式。超限输入被拒绝后仍保留已有撤销记录，图片 OCR 没有正文时保留原图发送方式。",
            "**问答卡片保持原位置**：回答后，交互问题不再被排到后续输出末尾。重新打开会话仍保留原有问答顺序。",
            "**取消后可以继续发送**：修复本地 OpenCode 取消影响下一次请求，以及回复完成后仍显示停止状态的问题。空闲队列不再反复闪动，子代理过程内容也不再被裁切。",
            "**工具错误可以纠正后继续**：远程任务在工具失败后已正常完成时，不再仍被误判为失败。内置 MCP 参数错误返回可纠正提示，笔记搜索单次最多支持 100 条结果。",
            "**菜单、通知和网页链接正常打开**：修复菜单及通知被模态窗口遮挡的问题。内置浏览器中需要新窗口的网页链接可在受管页签中打开。",
            "**知识图谱抽取不再额外限制为 8192 tokens**：OpenAI 协议请求由模型服务决定输出上限，Anthropic 沿用模型连接配置。服务商限制、超时与取消仍然有效。",
            "**桌面升级不再误拒绝旧 Agent 包**：签名、完整性和协议仍兼容的旧包，不会仅因其内置组件版本与桌面锁定版本不同而被拒绝。使用远程 Continue 等新能力仍需更新配套 Agent。"
          ],
          "notices": [
            "首次启动会升级数据库并将笔记正文迁移到本地文件，可能需要时间和额外磁盘空间。升级前请备份完整用户数据；旧客户端不能打开升级后的数据库，回退应使用升级前备份。",
            "后续笔记备份必须同时包含一致的数据库与相邻 `notes/` 目录。画布需手动保存，导出的 PDF 为栅格图像，不保留可搜索文字层。",
            "HTTP OCR 会把待识别内容发送到配置的服务，可能产生费用；高级选项效果取决于实际部署。快速文本／索引模式跳过 OCR，显式图片文字提取仍会调用所选 OCR。",
            "远程用户请先升级 **Desktop 0.13.8**，再下载并更新 **Agent 0.13.2**。远程项目仍为技术预览，Execute 使用所选 SSH 账号的权限。",
            "标准桌面包覆盖 Windows、macOS、Linux 的 x64／arm64，本次不构建龙芯实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Smoother settings scrolling**: Settings category titles and the active category tabs stay pinned at the top of the content area. Platform Features, Capabilities and Tools, message channels, and Runtime sub-tabs remain available while long forms scroll. Less frequently used controls stay in collapsed Advanced settings.",
            "**Copy model connections**: Copy an existing model connection from its detail view, including its endpoint, model, protocol, request customization, and authentication settings. Main copies a saved API key from encrypted credentials without exposing it to the UI or requiring re-entry.",
            "**Paged canvas notes**: Combine body text, handwriting, highlights, and images in Magic Notes, or import PDF pages for annotation. Manually save and resume editing, zoom from 25% to 300%, fit the width, and export to PDF.",
            "**Choose canvas pages for AI analysis**: Select the first 1–8 pages in note settings, with a default of one page and the same range for text and images. Saving and export retain the complete canvas of up to 50 pages; visual page analysis requires an image-capable model.",
            "**Connect HTTP PaddleOCR-VL**: Configure an OCR service in Document Parsing for both chat attachments and knowledge imports, without loading that service’s model weights locally. Test parsing with real files and configure layout, table, and other options supported by the service.",
            "**Keep and reuse document text and images**: Attachment originals and extracted images are retained while referenced, with text, image, and detail previews available after reopening conversations. Select document images into a named conversation draft or explicitly extract text from images; files already missing from older history are not recovered.",
            "**Manage built-in applications together**: Reorder four built-in applications from App Center, with matching order in the launcher and sidebar. Magic Notes and Local Inference Monitor have enablement and pinning settings, while Knowledge and Smart Heartbeat remain visible; a privately deployable application marketplace is still planned.",
            "**Browse notes and to-dos separately**: Note details show a continuous record stream with thumbnail navigation and independently collapsible index and AI-comment panes. The to-do view provides search, status filters, and details, preserving task position when returning from its source note.",
            "**View execution checklists in conversations**: Native OpenCode and Continue checklists appear above the conversation, expand to show progress, and persist with its history. The Runtime updates this read-only checklist; stopping a task does not mark unfinished items complete.",
            "**Use Continue in remote projects**: With the matching Agent update, managed SSH projects can select Continue for model requests, checklists, and conversation MCP tools in Execute. Cancellation, reconnection, and conversation continuation after restart are supported; Ask remains read-only.",
            "**Pin conversations and retain completion notices**: Pinned conversations remain pinned after restarting the app. Successful background conversations retain a completion notice during the current client session until their chat is opened.",
            "**Local inference monitoring and task statistics**: Inspect local service status and use each service’s supported stop, restart, or release actions; embedding shows independent-process CPU and memory, while speech and OCR retain shared-process measurement limitations. Task Center adds status counts and filters to find running tasks and those needing attention.",
            "**Read supplementary help on demand**: Page and settings explanations move into help controls beside titles or fields, accessible by mouse and keyboard. This reduces persistent explanatory text while keeping important operation consequences and errors visible."
          ],
          "fixes": [
            "**Run history migration only when needed**: The first launch after upgrading an older database still compacts repeated execution records. Once the database reaches the current schema, chatting and reopening GoodBuddy no longer show the history migration screen again.",
            "**Make model URL errors actionable**: Saving a model connection now identifies the specific connection when its endpoint is blank, too long, or unsupported, and keeps an accessible error beside the URL field instead of showing only a generic “Invalid URL”.",
            "**Keep older conversations and long histories accessible**: Conversation lists and search are no longer limited to the latest 100 conversations, and consecutive sends no longer truncate history beyond 500 messages. Model context limits still apply, and previously lost history cannot be recovered.",
            "**Preserve drafts and canvas edits**: Restoring queued text no longer overwrites newer input, and deleting canvas pages preserves body and checklist formatting. Rejected oversized input retains existing undo history, and image OCR without usable text preserves image delivery.",
            "**Keep question cards in their original position**: Answered interactive questions no longer move behind later output. Reopening the conversation preserves the question-and-answer order.",
            "**Continue sending after cancellation**: Fixed local OpenCode cancellation affecting the next request and completed replies retaining the Stop state. Idle queues no longer flicker repeatedly, and subagent progress is no longer clipped.",
            "**Recover from correctable tool errors**: Remote tasks that complete normally after a tool error are no longer incorrectly marked failed. Built-in MCP argument errors provide corrective guidance, and note searches support up to 100 results per request.",
            "**Open menus, notifications, and browser links correctly**: Fixed menus and notifications appearing behind modal windows. Browser links requesting a new window can open in managed tabs.",
            "**Remove the extra 8192-token graph-extraction limit**: OpenAI-protocol requests use provider output limits, while Anthropic follows the model connection setting. Provider limits, timeouts, and cancellation still apply.",
            "**Keep compatible older Agent packages usable**: Older packages with valid signatures, integrity, and compatible protocols are no longer rejected solely because bundled component versions differ from Desktop’s locks. New capabilities such as remote Continue still require the matching Agent update."
          ],
          "notices": [
            "First launch upgrades the database and moves note bodies into local files, which may take time and require additional disk space. Back up all user data first; older clients cannot open the upgraded database, so use a pre-upgrade backup for rollback.",
            "Subsequent note backups must include a consistent database and its adjacent `notes/` directory. Canvas notes require manual saving, and exported PDFs are raster images without a searchable text layer.",
            "HTTP OCR sends recognition content to the configured service and may incur charges; advanced-option results depend on the deployment. Fast text/index modes skip OCR, while explicit image text extraction still invokes the selected OCR.",
            "Remote users should upgrade to **Desktop 0.13.8** before downloading and updating to **Agent 0.13.2**. Remote projects remain a technical preview, and Execute uses the selected SSH account’s permissions.",
            "Standard desktop packages cover Windows, macOS, and Linux on x64/arm64, without a LoongArch experimental preview."
          ]
        }
      }
    },
    {
      "version": "0.13.6",
      "releasedAt": "2026-09-15",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**历史会话更省内存**：减少打开应用、刷新列表和切换会话时加载的历史数据，降低资源占用。历史详情按需读取，完整记录、全文搜索、复制、导出和续聊保持可用。",
            "**避免重复保存历史消息**：修复切回应用或刷新会话时，未变化的消息仍被反复保存的问题。减少重复处理，同时保留图片状态更新、后台审批和远程待答问题。"
          ],
          "notices": [
            "本次仅更新 Desktop，Agent 保持 **0.13.0**，无需因这两项修复重新更新 Host 环境。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Lower memory use with conversation history**: Reduced the history data loaded when opening the app, refreshing the list, and switching conversations. Details load on demand while complete records, full-text search, copy, export, and continued conversations remain available.",
            "**Avoid repeatedly saving unchanged messages**: Fixed unchanged messages being saved again when returning to the app or refreshing conversations. This reduces repeated processing while preserving image updates, background approvals, and pending remote questions."
          ],
          "notices": [
            "This release updates Desktop only. Agent remains at **0.13.0**, and these fixes do not require another Host environment update."
          ]
        }
      }
    },
    {
      "version": "0.13.5",
      "releasedAt": "2026-09-15",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**聊天生图，无需切换模型**：生图模块现在可作为 MCP 能力，供支持工具调用的聊天模型使用，无需另开会话或手动切换到图片模型。在当前对话中即可要求 AI 生成图片、修改上传图片或继续调整之前的成果，图片和生成状态直接显示在原会话中。",
            "**复制文件，直接粘贴为附件**：在文件管理器中复制一个或多个支持的本地文件后，直接粘贴到聊天输入框即可添加附件。读取和解析进度清晰可见，失败时保留草稿与已有附件，文本和截图仍可照常粘贴。"
          ],
          "fixes": [
            "**图片型 PPTX 也能读取文字**：配置本地 OCR 并选择启用 OCR 的解析方式后，可识别幻灯片内嵌图片中的文字，解决图片型 PPTX 无法提取内容的问题。识别结果保留幻灯片定位，原有文字也会一并保留。",
            "**识别结束后释放内存**：本地 OCR 在没有正在执行或排队的识别任务时，空闲 60 秒后自动释放模型运行内存。连续解析会复用模型，下次使用时自动重新加载。",
            "**多项目运行更省资源**：多个项目使用兼容配置时，本地 OpenCode、DeepSeek Harness 和托管远程 OpenCode 可分别复用各自的底层进程，减少重复启动和内存占用。不同会话仍可并行工作，停止其中一个不会连带停止其他会话。",
            "**远程对话恢复更连贯**：修复断线、桌面重启或配置更新后，远程待答问题和任务状态恢复不一致的问题。需要重试时，可直接在对应项目的菜单中操作。",
            "**项目菜单和工作栏更顺手**：长会话列表独立滚动，不再撑高项目菜单；没有会话活动时自动隐藏摘要。工作栏开关移到明暗主题按钮旁，不再悬浮遮挡内容。",
            "**魔法笔记自动同步显示**：AI 或 MCP 修改笔记后，已打开的笔记页面会自动刷新，无需手动重新打开。刷新保留当前选择和未保存草稿。",
            "**窗口异常时可以直接恢复**：界面渲染失败时提供“重新加载”入口，渲染进程崩溃时弹出恢复提示。重新加载可能丢失未保存的输入。"
          ],
          "notices": [
            "聊天生图需先为图片模型开启会话调用，并在 **Execute 模式**使用；支持直连文本模型、本地 OpenCode、Continue、DeepSeek Harness 和托管远程 OpenCode。生成和编辑可能产生图片服务费用，参考图编辑需要服务商支持。",
            "远程用户请先升级至 **GoodBuddy 0.13.5**，再更新 Host 上的 **Agent 0.13.0**。远程项目仍为技术预览，Ask/Execute 权限不变。",
            "提供 Windows、macOS、Linux 的 x64 和 arm64 标准桌面安装包，本次不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Create images without switching chat models**: Image generation is now available as an MCP capability to tool-capable chat models, without opening another conversation or manually switching to an image model. Ask AI to create an image, edit an upload, or refine an earlier result, with images and generation status shown in the same conversation.",
            "**Paste files directly into chat**: Copy one or more supported local files in your file manager and paste them into the composer to attach them. Reading and parsing progress is shown, failures preserve your draft and existing attachments, and text and screenshot paste continue to work."
          ],
          "fixes": [
            "**Read text from image-based PowerPoint files**: With local OCR configured and an OCR-enabled parsing mode selected, GoodBuddy can recognize text in embedded slide images instead of rejecting image-only PPTX files. Results retain slide locations alongside the original text.",
            "**Release memory after text recognition**: Local OCR releases its model memory after 60 idle seconds when no recognition tasks are running or queued. Consecutive tasks reuse the model, and it reloads automatically when needed again.",
            "**Use fewer resources across projects**: With compatible configurations, local OpenCode, DeepSeek Harness, and managed remote OpenCode each reuse their underlying processes to reduce duplicate startup and memory use. Conversations can still run in parallel, and stopping one does not stop the others.",
            "**Resume remote conversations more reliably**: Fixed inconsistent recovery of pending questions and task status after disconnection, Desktop restart, or configuration changes. Retry actions are available in the affected project’s menu.",
            "**Smoother project menus and workbar access**: Long conversation lists scroll independently without stretching the project menu, and the activity summary hides when idle. The workbar toggle now sits beside the theme switch instead of floating over content.",
            "**Keep Magic Notes up to date**: The open notes page refreshes automatically after AI or MCP edits, without needing to reopen it. Your selection and unsaved drafts are preserved.",
            "**Recover directly from window errors**: Rendering failures show a Reload action, and renderer-process crashes display a recovery prompt. Reloading may lose unsaved input."
          ],
          "notices": [
            "Enable conversation access for an image model and use **Execute mode** for in-chat image tools. Supported runtimes are direct text models, local OpenCode, Continue, DeepSeek Harness, and managed remote OpenCode; image requests may incur provider charges, and reference-image editing requires provider support.",
            "Remote users should upgrade to **GoodBuddy 0.13.5** before updating Hosts to **Agent 0.13.0**. Remote projects remain a technical preview, with unchanged Ask/Execute permissions.",
            "Standard desktop packages cover Windows, macOS, and Linux on x64 and arm64. No LoongArch experimental preview is included."
          ]
        }
      }
    },
    {
      "version": "0.13.2",
      "releasedAt": "2026-09-14",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**修复升级后仍重复增长的执行记录**：修复 0.13.1 中子代理使用工具时，未变化的工具内容仍被重复保存的问题。首次启动会无损整理这些重复内容，保留会话、执行详情、结果和远程来源记录。任务结束后及时释放存储缓存，不清空其他正在运行会话的缓存。",
            "**工具正在做什么更清楚**：OpenCode、Continue、DeepSeek Harness 和直连模型的工具记录可显示文件路径、命令或搜索摘要。工具完成后仍保留已有操作说明，展开记录或重新打开会话可继续查看输入、结果和错误。",
            "**修复 Windows DeepSeek 工作区路径错误**：同一个目录使用不同分隔符或规范路径写法时，不再因为 Host 与会话路径不一致而导致请求失败。Host 启动和后续会话使用同一规范路径，保持原有工作区与权限边界。",
            "**跨项目活动菜单更紧凑**：点击活动摘要后，可按“项目 → 会话”两级菜单直接进入需要处理的对话，不再打开整块活动弹窗。菜单支持键盘导航，并调整了入口间距和会话行指示，保留原有活动计数及跨项目定位。"
          ],
          "notices": [
            "本次会再次整理已升级至 0.13.1 的数据库；大库可能需要时间和额外磁盘空间，支持退出后继续。升级后的数据库不能由旧客户端打开，需要回退时请使用升级前备份，不要手工修改数据库版本。",
            "配套发布 **Agent 0.11.25** 维护版，需要先升级至 **Desktop 0.13.2**，再按需更新 Host 环境。存储补修由 Desktop 执行，不依赖升级 Agent；远程项目仍为技术预览，Ask/Execute 权限不变。",
            "Desktop 提供 Windows、macOS、Linux 的 x64/arm64 标准安装包；本次不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Fix repeated execution-history growth after upgrading**: Fixed an issue in 0.13.1 where subagents using tools could repeatedly save unchanged tool content. The first launch compacts these duplicates without losing conversations, execution details, results, or remote provenance. Finished tasks release their storage caches without clearing caches for other active conversations.",
            "**See what tools are doing**: Tool records from OpenCode, Continue, DeepSeek Harness, and direct models can show file paths, commands, or search summaries. Existing operation descriptions remain after completion, and expanded records or reopened conversations retain inputs, results, and errors.",
            "**Fix Windows DeepSeek workspace-path errors**: Equivalent paths to the same directory no longer fail because the Host and session disagree about separators or canonical spelling. Host startup and subsequent sessions use the same canonical path while preserving workspace and permission boundaries.",
            "**More compact cross-project activity navigation**: Open the activity summary and use a two-level project-to-conversation menu to reach conversations needing attention instead of opening a large activity dialog. Keyboard navigation, improved spacing, and simpler conversation indicators preserve activity counts and exact cross-project navigation."
          ],
          "notices": [
            "This update also compacts databases already upgraded to 0.13.1; large databases may take time and require extra disk space, with quit-and-resume support. Older clients cannot open the upgraded database, so use a pre-upgrade backup for rollback rather than manually changing its version.",
            "The accompanying **Agent 0.11.25** maintenance release requires **Desktop 0.13.2** before updating a Host environment. Desktop performs the storage repair without requiring an Agent update; remote projects remain a technical preview and Ask/Execute permissions are unchanged.",
            "Desktop includes standard Windows, macOS, and Linux x64/arm64 packages, without a LoongArch experimental preview."
          ]
        }
      }
    },
    {
      "version": "0.13.1",
      "releasedAt": "2026-09-13",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**修复数据库异常膨胀**：长任务和子任务不再反复保存整份执行过程。首次启动自动整理旧记录并回收磁盘空间，保留聊天、执行详情和结果。",
            "**浏览器不再遮挡弹窗**：打开项目设置、知识库实例等弹窗时，内置网页会正确让出显示区域。关闭弹窗后恢复原网页，不丢失页面状态。",
            "**知识库配置更清楚**：长名称和长参数表单不再挤压来源选择控件。切换远端知识库时自动名称随之更新，手工填写的名称保持不变。",
            "**更新说明恢复显示**：修复历史更新说明增长后，本次更新内容可能无法加载的问题。当前版本说明可正常显示并确认，重启后不再重复弹出。"
          ],
          "notices": [
            "整理较大的历史数据库可能需要较长时间和额外磁盘空间，支持退出后下次继续。升级后的数据库不能由旧版客户端打开，需要回退时请保留升级前备份。",
            "Agent 保持 **0.11.24**，本次不重新发布 Agent。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Fix excessive database growth**: Long-running tasks and subagents no longer repeatedly save the entire execution history. The first launch compacts existing records and reclaims disk space while preserving chats, execution details, and results.",
            "**Keep dialogs above embedded webpages**: Embedded webpages now yield their display area when project settings, knowledge-instance dialogs, and other overlays open. Closing the overlay restores the original webpage without losing its state.",
            "**Clearer knowledge configuration**: Long names and parameter forms no longer squeeze source-selection controls. Automatically filled names follow the selected remote knowledge base, while manually entered names are preserved.",
            "**Restore release-note display**: Fixed an issue where growing historical release notes could prevent the current update notes from loading. Current-version notes can be displayed and acknowledged without appearing again after restart."
          ],
          "notices": [
            "Large history databases may require substantial processing time and extra disk space; you can quit and continue on the next launch. Older clients cannot open the upgraded database, so retain a pre-upgrade backup if rollback is needed.",
            "Agent remains at **0.11.24** and is not republished."
          ]
        }
      }
    },
    {
      "version": "0.13.0",
      "releasedAt": "2026-09-13",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**接入外部知识库**：在知识库页面配置 Dify、FastGPT 或 RAGFlow 实例，从目录选择或手工填写远端知识库 ID，验证后绑定到 GoodBuddy。对话可按需或每次回答前检索，并保留可查看的引用片段。连接、参数和移除操作只影响 GoodBuddy，不同步、索引或修改远端内容。",
            "**跨项目查看活动**：无需逐个切换项目，即可集中查看正在运行和需要处理的对话。点击活动可直接进入对应会话，项目选择器也会显示活动数量。"
          ],
          "fixes": [
            "**远程 OpenCode 原生问答**：托管 SSH 前台对话可以接收主会话及子会话的问题，通过选项、自由文本或跳过继续执行。需要配套 Agent `0.11.24`，不扩展到任意第三方 ACP 服务。",
            "**并行提问不再相互覆盖**：同时到达的问题按顺序等待回答，重复事件不会清空草稿，提交失败可保留输入重试。配套更新 Agent 后，取消远程待答任务也可在同一会话继续发送，不再因旧操作绑定报错。",
            "**工具记录更紧凑**：工具调用改为可展开的简洁记录，减少重复摘要。展开后可查看和复制结果、错误及输入参数。",
            "**输入区与搜索更清楚**：会话选项收进紧凑设置面板，保留 Runtime、工作模式、发送和队列操作。搜索框可一键清除关键词，返回当前范围的会话列表。",
            "**浏览器关闭与刷新更可靠**：AI 正在使用的页签也可关闭，不会因此取消整个请求；后续显式导航会创建独立新页。空白页不再错误启用刷新，切换或关闭面板后也不会残留旧错误。",
            "**弹窗按钮恢复正常点击**：修复项目设置、反馈、知识库、MCP 编辑器和预览等弹窗与标题栏重叠时，控件无法悬停或点击的问题。项目对话框滚动后仍可关闭，关闭时恢复入口焦点。",
            "**文件读取与搜索错误可继续处理**：直连模型遇到可修正的路径、读取或搜索参数错误时，可依据错误提示调整后继续。部分文件无法搜索时保留已有结果并明确提示覆盖不完整，不把缺失结果当作不存在。"
          ],
          "notices": [
            "外部检索会把查询发送给所选知识服务，回答使用的片段会进入模型上下文并随会话保存在本机。知识服务的向量化、重排及模型回答可能产生服务商用量。",
            "先升级 Desktop 至 **0.13.0**，再下载并更新 Host 上的 **Agent 0.11.24**；远程项目仍为技术预览，Ask/Execute 权限不变。本次提供标准六平台桌面安装包，不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Connect external knowledge bases**: Configure Dify, FastGPT, or RAGFlow on the Knowledge page, select a remote knowledge base from its catalog or enter its ID, then verify and bind it to GoodBuddy. Conversations can retrieve on demand or before every answer and retain readable citation snippets. Connection settings, retrieval parameters, and removal affect only GoodBuddy, without synchronizing, indexing, or modifying remote content.",
            "**View activity across projects**: See running conversations and conversations needing attention without switching through projects individually. Open a conversation directly from its activity entry, with activity counts also shown in the project selector."
          ],
          "fixes": [
            "**Native questions in remote OpenCode**: Foreground managed SSH conversations can receive questions from the main session and its child sessions, then continue through choices, free-text answers, or skipping. This requires Agent `0.11.24` and does not add support for arbitrary third-party ACP services.",
            "**Concurrent questions no longer overwrite each other**: Questions arriving together wait in order, duplicate events preserve drafts, and failed submissions retain input for retry. With the matching Agent update, cancelling a pending remote question also allows another message in the same conversation without an old-operation binding error.",
            "**More compact tool records**: Tool calls use concise expandable records with fewer repeated summaries. Expand a record to inspect and copy results, errors, and input parameters.",
            "**Clearer composer and search**: Conversation options move into a compact settings panel while preserving Runtime, work mode, sending, and queue actions. Clear search keywords in one click to return to conversations in the current scope.",
            "**More reliable browser closing and refresh**: Tabs in use by an AI request can be closed without cancelling the whole request; a later explicit navigation creates a separate replacement tab. Blank pages no longer incorrectly enable refresh, and switching or closing panels clears stale errors.",
            "**Clickable dialog controls**: Fixed controls that could not respond to hover or clicks when project settings, feedback, knowledge, MCP editors, and previews overlapped the titlebar. Project dialogs remain closable after scrolling and restore focus to their entry point.",
            "**Recoverable file-reading and search errors**: Direct models can correct actionable path, reading, or search-parameter errors and continue. When some files cannot be searched, available results are retained with an explicit incomplete-coverage warning rather than treating missing results as proof of absence."
          ],
          "notices": [
            "External retrieval sends queries to the selected knowledge service, and snippets used in answers enter model context and are saved locally with the conversation. Embedding, reranking, and model answers may incur provider usage charges.",
            "Upgrade Desktop to **0.13.0** before downloading and updating Hosts to **Agent 0.11.24**. Remote projects remain a technical preview, Ask/Execute permissions are unchanged, and this release includes the six standard desktop targets without a LoongArch experimental preview."
          ]
        }
      }
    },
    {
      "version": "0.12.12",
      "releasedAt": "2026-09-11",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**连续改图保留上下文**：继续修改图片时，自动带上最近一次成功生成的图片与文字历史，重新打开已保存会话后仍可接着修改，本轮显式附图优先。上游明确不支持图片编辑时继续按文字生成，并在回复底部说明未使用参考图片；真实服务错误和取消仍正常显示。",
            "**提交后仍可查看问题与回答**：OpenCode 等前台 Runtime 问答在回答或跳过成功后保留原问题与对应答案，多轮记录随本地会话保存。提交失败仍可重试，较晚返回的提交结果不会覆盖下一轮问题或任务终态。",
            "**Runtime 与模型选择更清楚**：会话和项目菜单不再重复列出每个 Runtime 与模型的组合，需要调整 Runtime 模型时请进入系统设置。已保存的项目固定模型继续有效，帮助文字也会明确显示固定模型，而不是误报为跟随全局配置。",
            "**Git 工作区切换与历史浏览更顺畅**：切换项目后自动识别 Git 工作区，不再需要手动刷新，也不会混入上一个项目的状态。提交历史默认贴底，首次展开占用一半内容区，支持拖动或键盘调整比例，上下列表独立滚动并统一使用顶部刷新。"
          ],
          "notices": [
            "旧版本未保存的问题与答案无法恢复。本次不改变 Ask/Execute 权限，也不新增远程 ACP 问答能力。",
            "Agent 保持 **0.11.23**，本次不重新发布 Agent。提供 Windows、macOS、Linux 的 x64/arm64 标准安装包，不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Keep context across image edits**: Follow-up image requests automatically include the latest successfully generated image and text history, even after reopening a saved conversation, while explicit attachments take priority. If the provider explicitly does not support image editing, generation continues from text with a footer explaining that reference images were not used; genuine service errors and cancellation remain visible.",
            "**Review questions and answers after submission**: Foreground Runtime questions, including OpenCode questions, retain their original questions and answers after a successful answer or skip, with multiple rounds saved in the local conversation. Failed submissions remain retryable, and late submission results do not overwrite a newer question or the task's terminal state.",
            "**Clearer Runtime and model selection**: Conversation and project menus no longer repeat every Runtime/model combination; change Runtime models in system settings. Saved fixed project models remain effective, and their help text explicitly identifies the fixed model instead of incorrectly claiming to follow global settings.",
            "**Smoother Git project switching and history browsing**: Switching projects automatically detects the Git workspace without a manual refresh or another project's stale status. Commit history stays docked at the bottom, initially expands to half the content area, supports pointer and keyboard resizing, and uses independently scrolling panes with one shared refresh action."
          ],
          "notices": [
            "Questions and answers not saved by older versions cannot be recovered. This release does not change Ask/Execute permissions or add remote ACP question support.",
            "Agent remains at **0.11.23** and is not republished. Standard Windows, macOS, and Linux x64/arm64 packages are included, without a LoongArch experimental preview."
          ]
        }
      }
    },
    {
      "version": "0.12.11",
      "releasedAt": "2026-09-10",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**工作栏管理项目文件与 Git**：可创建、重命名、移动和删除文件或目录，并查看属性、分支、提交历史和单文件差异。文件与 Git 工具按视图分组，分支搜索统一样式，子目录保留文字导航。",
            "**多个独立浏览器页签**：可同时打开多个浏览器页签，分别保留页面、导航和交互状态，并支持全屏显示。同一对话的页签共享登录状态，模型请求固定使用开始时绑定的页签。",
            "**直连模型开发工具**：可使用随包 ripgrep 搜索项目文件与内容，并分页读取大型文本。Execute 可应用多文件补丁，Ask 仍保持只读。",
            "**任务中心范围切换**：任务中心可查看当前项目、全局任务或全部项目。所选范围随工作栏布局保存，任务中心和工作区保留为基础页签。"
          ],
          "fixes": [
            "**长上下文与工具输出不再过早截断**：移除历史消息、长回复和文档解析文本的旧固定截断限制，并支持分页续读命令与 Subagent 输出。所选模型的上下文窗口仍然有效。",
            "**定时消息沿用会话设置**：定时消息使用会话当前历史、Runtime、工作模式和知识检索设置。排队、运行和实际工作模式及时显示，同一计划尚未完成时不能重复“立即运行”。",
            "**模型选择与用量保持一致**：未固定的连接继续跟随当前兼容默认值，显式选择保持固定。手动压缩、远程会话用量和 Harness 来源显示与实际配置保持一致。",
            "**浏览器显示与恢复更可靠**：修复 Agent 页面与工作栏页签对应、已释放页面恢复以及跨会话选择的问题。未使用浏览器时不创建后台页面，普通操作不再触发无用的自动截图。",
            "**文件操作与工具调用更顺畅**：修复目录和提交历史刷新、仅修改文件名大小写，以及重命名文件的提交差异。健康的 MCP 调用复用已发现工具，减少重复发现带来的等待。"
          ],
          "notices": [
            "远程新增文件/Git 操作及长任务修复需要 **Agent 0.11.23**。请先升级 Desktop 至 **0.12.11**，再下载并更新 Host 上的 Agent。",
            "删除操作经确认后永久执行，移动限于工作区内，分支切换不会自动暂存或丢弃冲突改动。Ask 保持只读，默认 Execute 自动授权已启用工具；本次不构建 LoongArch 预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Manage project files and Git from the workbar**: Create, rename, move, and delete files or directories, and inspect properties, branches, commit history, and per-file diffs. File and Git actions are grouped by view, branch search uses consistent styling, and subdirectories retain text breadcrumbs.",
            "**Multiple independent browser tabs**: Open several browser tabs with separate pages, navigation, and interaction state, including fullscreen display. Tabs in one conversation share login state, while each model request retains its starting tab binding.",
            "**Direct-model development tools**: Use packaged ripgrep to search project files and content, and read large text files in pages. Execute can apply multi-file patches, while Ask remains read-only.",
            "**Task Center scopes**: View tasks for the current project, global tasks, or all projects. The selected scope is saved with the workbar layout, while Task Center and Workspace remain permanent tabs."
          ],
          "fixes": [
            "**Prevent premature context and tool-output truncation**: Removed previous fixed truncation limits on message history, long replies, and parsed documents, with paged access to command and Subagent output. The selected model’s context window still applies.",
            "**Scheduled messages retain conversation settings**: Scheduled messages use the conversation’s current history, Runtime, work mode, and knowledge retrieval settings. Queue state, execution state, and actual work mode stay current, and Run Now cannot duplicate an unfinished occurrence.",
            "**Consistent model selection and usage**: Unpinned connections follow the compatible current default, while explicit selections remain fixed. Manual compaction, remote conversation usage, and the displayed Harness source reflect the actual configuration.",
            "**More reliable browser display and recovery**: Fixed Agent-to-workbar tab mapping, released-page recovery, and explicit cross-conversation selection. Unused browser capabilities do not create background pages, and ordinary operations no longer trigger unused automatic screenshots.",
            "**Smoother file operations and tool calls**: Fixed directory and commit-history refresh, case-only filename changes, and commit diffs for renamed files. Healthy MCP calls reuse discovered tools to reduce repeated discovery delays."
          ],
          "notices": [
            "New remote file/Git operations and long-task fixes require **Agent 0.11.23**. Upgrade Desktop to **0.12.11** before downloading and updating the Host’s Agent.",
            "Confirmed deletion is permanent, moves stay inside the workspace, and branch switching never automatically stashes or discards conflicts. Ask remains read-only, Execute automatically authorizes enabled tools by default, and this release excludes the LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.12.10",
      "releasedAt": "2026-09-09",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**OpenCode 并行执行更顺畅**：修复批量改写文件或转换换行符后，原生自动 Git 快照计算可能拖住其他会话的问题。GoodBuddy 托管的本机与远端 OpenCode 不再生成界面未使用的自动快照，文件工具、子代理和工作区 Git 差异仍然可用。",
            "**Continue 输出不再堵住宿主**：修复控制台输出较多时可能填满子进程管道、使请求无法继续的问题。现在会持续排空不用展示的输出，正常回复、工具进度和取消仍沿用原有通道。",
            "**聊天内容与输入框对齐**：头像移到消息正文外侧，消息边缘与输入框保持一致。滚动条空间留在外侧，避免聊天区域出现多余的横向滚动。",
            "**待办来源就在详情里**：魔法笔记把来源记录放入所选待办的详情区域，查看事项和原文不必再找底部面板。切换待办时不会混入上一项的来源内容，读取失败通过应用通知提示。",
            "**便携包构建**：修复 Windows 打包准备 Runtime 时依赖特定 npm 环境变量的问题。构建流程现在也能使用项目内 npm 或系统 npm，避免因此中断便携包构建。"
          ],
          "notices": [
            "本机修复随升级并重启 Desktop 生效；远程项目请先升级 Desktop 至 0.12.10，再更新 Agent 至 0.11.22。已运行的旧 Runtime 不会热更新，外部 OpenCode Server 的配置不受此调整影响。",
            "本次使用 Windows、macOS、Linux 的 x64/arm64 六平台桌面发布矩阵，不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Smoother parallel OpenCode execution**: Fixed automatic native Git snapshot calculations that could stall other conversations after broad file rewrites or line-ending conversions. GoodBuddy-managed local and remote OpenCode no longer generate snapshots unused by the interface; file tools, subagents, and workspace Git diffs remain available.",
            "**Continue output no longer blocks its host**: Fixed larger console output filling a child-process pipe and preventing requests from progressing. Unused output is now continuously drained, while replies, tool progress, and cancellation keep using their existing channels.",
            "**Aligned messages and composer**: Avatars now sit outside message content, with message edges aligned to the composer. Scrollbar space stays in the outer gutter to avoid unnecessary horizontal scrolling in the chat area.",
            "**Todo sources inside their details**: Magic Notes now shows the source entry within the selected todo's detail area instead of a separate bottom panel. Switching todos cannot show the previous item's source content, and failed reads use application notifications.",
            "**Portable package builds**: Fixed Windows Runtime preparation depending on a particular npm environment variable. Builds can now also use the project's bundled npm or the system npm instead of failing portable packaging for that reason."
          ],
          "notices": [
            "Local fixes take effect after upgrading and restarting Desktop. For remote projects, upgrade Desktop to 0.12.10 before updating Agent to 0.11.22. Already-running Runtimes are not hot-updated, and external OpenCode Server configuration is unchanged.",
            "This release uses the six standard Windows, macOS, and Linux x64/arm64 desktop targets, without a LoongArch experimental preview."
          ]
        }
      }
    },
    {
      "version": "0.12.9",
      "releasedAt": "2026-09-08",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**待办来源记录**：在魔法笔记的待办视图选择事项后，可直接查看对应的来源记录。遇到读取失败可以重试，不必来回切换查找笔记。"
          ],
          "fixes": [
            "**消息状态更准确**：回复底部会区分准备请求、等待重试、开始重试、工具活动和等待回答，本机 OpenCode 还会显示原生重试次数与计划时间。没有新进展时不再用闪烁暗示模型仍在工作，会话列表闪点及发送、停止按钮保持不变。",
            "**并行 OpenCode 会话**：修复开启另一条本机会话时，已有会话可能一直等待的问题。服务异常退出时会明确提示失败，不会自动重复执行已运行的命令。",
            "**参考图编辑**：修复部分兼容图像服务商拒绝图片编辑请求的问题。使用参考图修改图片时，不再附带只适用于生成接口的默认参数。",
            "**原生问答与长上下文**：修复 OpenCode、Continue 的较长问题、较多选项或较长自定义回答被截断或拒绝的问题，并提高 Continue 的请求正文容量。用户可以继续通过现有问答卡片回答，较长上下文不再受旧的小容量限制阻断。",
            "**推理排版**：修复 OpenCode 流式推理片段拼接时丢失 Markdown 分段的问题。推理过程中的段落和换行会按原始内容保留。",
            "**运行记录提示**：移除常驻的旧版活动历史截断警告，减少与当前执行无关的提示。此调整不会恢复旧版本已经丢失的历史记录。"
          ],
          "notices": []
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Todo source entries**: Select an item in the Magic Notes todo view to read its source entry directly. Failed reads can be retried without switching back and forth to find the original note."
          ],
          "fixes": [
            "**More accurate message status**: The reply footer distinguishes request preparation, retry waiting, retry dispatch, tool activity, and pending answers; local OpenCode also reports native retry attempts and scheduled times. The footer no longer pulses to imply progress when none is reported, while conversation-list indicators and send/stop buttons remain unchanged.",
            "**Parallel OpenCode conversations**: Fixed an issue where opening another local conversation could leave an existing one waiting indefinitely. Unexpected server exits now produce an explicit failure without automatically replaying commands that already ran.",
            "**Reference-image editing**: Fixed editing requests rejected by some compatible image providers. Reference-image edits no longer include default parameters that belong only to the generation endpoint.",
            "**Native questions and longer context**: Fixed longer questions, larger option lists, and longer custom answers being truncated or rejected in OpenCode and Continue, and increased Continue's request-body capacity. Users can keep answering through the existing question card without longer context hitting the previous small request limit.",
            "**Reasoning formatting**: Fixed lost Markdown boundaries when streamed OpenCode reasoning fragments were joined. Paragraphs and line breaks now retain their original structure.",
            "**Run-history notices**: Removed the persistent legacy activity-history truncation warning to reduce notices unrelated to current work. This does not restore history already lost in older versions."
          ],
          "notices": []
        }
      }
    },
    {
      "version": "0.12.8",
      "releasedAt": "2026-09-08",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**修复 OpenCode 启动失败。** 安装包补齐离线配置依赖，避免因缺少 `@opencode-ai/plugin/package.json` 无法启动。升级后无需手动安装 Node.js 或 npm 依赖。",
            "**修复 DeepSeek Harness Host 启动退出。** 将必需的会话组件完整打入 Host，避免安装后提示“Host 启动前退出（code 1）”。现有模型配置保持不变。",
            "**macOS 仅提供 DMG。** Intel 与 Apple 芯片版本不再构建或发布 ZIP。Windows 和 Linux 的安装包格式保持不变。"
          ],
          "notices": [
            "旧版 macOS 的 GitHub 更新检查，以及所有平台的旧版镜像更新检查可能无法识别新清单，请从官方下载页手动升级至 `0.12.8`。",
            "Agent 继续使用 `0.11.21`，本次不重发 Agent，也不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Fixed OpenCode startup failures.** Desktop packages now include the offline configuration dependencies, avoiding startup failures caused by a missing `@opencode-ai/plugin/package.json`. No manual Node.js or npm dependency installation is needed after upgrading.",
            "**Fixed DeepSeek Harness Host startup exits.** Required session components are bundled into the Host, preventing the “Host exited before startup (code 1)” error after installation. Existing model settings remain unchanged.",
            "**DMG-only macOS releases.** Intel and Apple silicon versions no longer build or publish ZIP packages. Windows and Linux package formats are unchanged."
          ],
          "notices": [
            "Older macOS GitHub update checks and older mirror update checks on all platforms may reject the new manifest. Upgrade manually to `0.12.8` from the official download page.",
            "Agent remains at `0.11.21`; this release does not republish the Agent or include a LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.12.7",
      "releasedAt": "2026-09-08",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**参考图编辑。** 使用兼容的图像模型连接时，可以附加参考图片并描述修改要求。GoodBuddy 会发送图片编辑请求；服务商必须支持图片编辑接口。",
            "**专家协作沿用工作模式。** 选择专家或团队分析时，专家可以按当前 Ask/Execute 模式使用已启用的本机直连模型工具。Ask 仍只读；这些工具不是远程 OpenCode 子会话。"
          ],
          "fixes": [
            "**Execute 不再卡在目录权限等待。** OpenCode、Continue 和 DeepSeek Harness 可以按当前账号权限访问工作区外目录。OpenCode 原生子代理的权限请求也会正确处理，Ask 保持只读。",
            "**原生问答正确回传。** 本机 OpenCode 和 Continue 的选项、yes/no、自由文本及跳过通过现有问答卡片处理。修复 OpenCode 子会话提问丢失，以及 Continue 自定义答案回传错误。",
            "**OpenCode 会话切换更稳定。** 切回会话时保留可用的原生操作栏，并在复用服务前检查其状态。并行请求的取消和临时 MCP 清理保持隔离。",
            "**Runtime 启动更可靠。** OpenCode 使用随包提供并按版本复用的离线配置依赖，减少重复准备。修复 DeepSeek Harness 升级后的初始化失败。",
            "**远程 Agent 安装与重连更可靠。** 修复包含 OpenCode 离线依赖的安装包被误判为元数据过大，以及部分合法握手被拒绝的问题。连接在握手期间关闭时会及时返回错误，不再一直等待。",
            "**聊天输入与布局更顺畅。** 输入时不再重复渲染未变化的聊天历史，消息正文统一显示在头像和名称下方。终端移除多余内边距，让内容使用完整面板空间。"
          ],
          "notices": [
            "先升级至 Desktop `0.12.7`，再安装 Agent `0.11.21`，以获得远程修复。远程项目仍为技术预览；Agent 单独提供 Linux x64、Linux arm64 和 macOS arm64 包。",
            "OpenCode 更新至 `1.18.29`；已运行的旧 Runtime 不会热更新，需要升级并重启。结构化业务问题仍需用户回答，不会自动代答终端密码、OAuth 或任意确认提示。",
            "本次使用标准六平台桌面发布矩阵，不构建 LoongArch 实验预览。应用导航与系统工具入口文档仍为设计稿，不代表新界面已经交付。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Reference-image editing.** Attach reference images and describe the changes when using a compatible image-model connection. GoodBuddy sends an image-edit request; the provider must support the image-editing endpoint.",
            "**Experts inherit the work mode.** Selected experts and team analysis can use enabled local direct-model tools under the current Ask/Execute mode. Ask remains read-only; these tools are not remote OpenCode child sessions."
          ],
          "fixes": [
            "**Execute no longer stalls on directory permissions.** OpenCode, Continue, and DeepSeek Harness can access paths outside the workspace using the current account’s permissions. OpenCode native subagent permission requests are also handled correctly, while Ask stays read-only.",
            "**Native questions return answers correctly.** Local OpenCode and Continue handle choices, yes/no, free text, and skipping through the existing question cards. This fixes lost OpenCode child-session questions and incorrect Continue custom-answer responses.",
            "**More stable OpenCode session switching.** Returning to a conversation preserves usable native controls, and cached services are checked before reuse. Cancellation and temporary MCP cleanup remain isolated between parallel requests.",
            "**More reliable Runtime startup.** OpenCode uses bundled offline configuration dependencies and reuses them by version, reducing repeated preparation. DeepSeek Harness initialization failures after upgrading are also fixed.",
            "**More reliable remote Agent installation and reconnection.** Packages containing OpenCode’s offline dependencies are no longer incorrectly rejected as oversized metadata, and valid handshakes are no longer intermittently rejected. Connections closed during the handshake now return an error promptly instead of waiting indefinitely.",
            "**Smoother typing and cleaner layouts.** Typing no longer rerenders unchanged chat history, and message content appears below the avatar and name. Removing extra terminal padding gives content the full panel area."
          ],
          "notices": [
            "Upgrade to Desktop `0.12.7` before installing Agent `0.11.21` for the remote fixes. Remote Projects remain a technical preview; separate Agent packages target Linux x64, Linux arm64, and macOS arm64.",
            "OpenCode updates to `1.18.29`; running older Runtimes require an upgrade and restart rather than a live update. Structured business questions still require user answers; terminal passwords, OAuth, and arbitrary confirmation prompts are not answered automatically.",
            "This release uses the standard six-target desktop matrix without a LoongArch preview. Application navigation and system-tool entry documents remain design drafts, not delivered UI."
          ]
        }
      }
    },
    {
      "version": "0.12.6",
      "releasedAt": "2026-09-06",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**子代理过程与结果显示更完整。** 展开 OpenCode 子代理卡片，可以按顺序查看文字、推理和工具详情，最终结果单独展示。远程项目需要更新 Agent 才能查看实时过程；旧 Agent 仍可显示最终结果，但不会补造缺失的过程记录。",
            "**工作区文件变更更容易查看。** 点击变更文件，可分别查看已暂存和未暂存差异，包括删除、重命名和未跟踪文件。刷新会更新当前差异和已展开目录，超过 50 项的变更可以继续加载。",
            "**知识库支持整行勾选。** 点击知识库名称或所在行即可切换选择，不再因焦点变化导致点击无效。选择范围仍按当前对话保存。",
            "**OpenCode 连续与并行对话更稳定。** 每次请求结束后会释放自身的事件连接，避免连续使用后连接不断积累。取消一条请求不会关闭其他并行会话的连接。",
            "**命令可以使用工作区外目录。** 直连模型在 Execute 模式下，可以使用绝对路径、相对路径或符号链接指定工作目录。未指定时仍使用项目工作区，Ask 保持只读。",
            "**恢复 GitHub 来源的 Agent 更新检查。** 选择 GitHub 后，可以正常读取可用 Agent 版本，不再因历史发布信息过大提示“发布目录超出大小限制”。镜像入口和已有本地包不受影响。"
          ],
          "notices": [
            "远程子代理实时过程和本轮远程文件差异改进需要更新至 Agent `0.11.20`。提供 Linux x64、Linux arm64 和 macOS arm64 包，Desktop 不内置 Agent。",
            "本次使用标准六平台桌面发布矩阵，不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**More complete subagent progress and results.** Expand an OpenCode subagent card to view ordered text, reasoning, and tool details, with the final result shown separately. Remote projects require an updated Agent for live progress; older Agents still show final results without inventing missing progress history.",
            "**Clearer workspace file changes.** Select a changed file to inspect staged and unstaged diffs separately, including deleted, renamed, and untracked files. Refresh updates the selected diff and expanded directories, and lists with more than 50 changes can load additional entries.",
            "**Whole-row knowledge selection.** Click a knowledge library’s name or row to toggle its selection without focus changes losing the click. Selection remains scoped to the current conversation.",
            "**More reliable consecutive and parallel OpenCode conversations.** Each request releases its own event connection when it ends, preventing connections from accumulating during continued use. Cancelling one request does not close other parallel conversations’ connections.",
            "**Commands can use directories outside the workspace.** Direct models in Execute mode can select a working directory using an absolute path, a relative path, or a symbolic link. The project workspace remains the default, and Ask stays read-only.",
            "**Restored Agent update checks through GitHub.** Selecting GitHub now retrieves available Agent versions without failing because the release history exceeds the response-size limit. Mirror access and existing local packages are unaffected."
          ],
          "notices": [
            "Update to Agent `0.11.20` for live remote subagent progress and the remote file-diff improvements in this release. Packages are available for Linux x64, Linux arm64, and macOS arm64; Desktop does not embed the Agent.",
            "This release uses the standard six-target desktop matrix and does not include an experimental LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.12.5",
      "releasedAt": "2026-09-06",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [
            "**Apple Silicon Mac 远程项目。** 可以把 macOS arm64 设备添加为 SSH Host，安装独立 Agent 后使用 Ask、Execute 和断线重连。无需管理员权限或 launchd 服务；Intel Mac 暂不支持，远程项目仍为默认关闭的技术预览。"
          ],
          "fixes": [
            "**知识库使用流程更完整。** 创建和检索测试默认收起高级参数，资料可用后可直接“用于当前对话”；知识范围按对话保存，新对话默认不选择知识库。文档明确区分可用、处理中和失败，支持打开来源、失败重试，并在移除来源前说明影响和要求确认。",
            "**切换窗口不再误中断回复。** 来回切换桌面窗口时，仍在生成的回复不会被误判为“上次运行意外中断”，后续消息也不会因此滞留队列。OpenCode 专属操作栏在会话刷新时保持稳定，减少界面抖动。",
            "**长会话可以完整滚动。** 较长推理、工具记录或正文不再因估算高度导致滚动条提前到底。全部已加载内容可直接滚动查看，无需切换会话刷新。"
          ],
          "notices": [
            "**先升级桌面端，再更新 Agent。** Agent `0.11.19` 提供 Linux x64、Linux arm64 和 macOS arm64 包，共用一个签名目录。旧桌面端无法解析包含 macOS 条目的新目录，需升级至 Desktop `0.12.5` 后检查在线 Agent 更新；已有本地包和历史发布仍保留。",
            "Desktop 不内置 Agent 或远程 Runtime。Ask 保持只读，Execute 使用所选 SSH 账号的完整权限。",
            "本次使用标准六平台桌面发布矩阵，不构建 LoongArch 实验预览。",
            "仓库新增独立 ShareServer 管理控制台交互原型，不包含在桌面安装包中，也不是已上线的团队共享服务；原型数据操作不持久化。"
          ]
        },
        "en-US": {
          "highlights": [],
          "features": [
            "**Remote projects on Apple Silicon Macs.** Add a macOS arm64 device as an SSH Host and install the separate Agent to use Ask, Execute, and reconnection. No administrator access or launchd service is required; Intel Macs are not supported, and Remote Projects remain an off-by-default technical preview."
          ],
          "fixes": [
            "**A more complete knowledge workflow.** Creation and retrieval testing collapse advanced settings by default, and searchable libraries can be sent directly to the current conversation. Knowledge selection is saved per conversation and starts empty for new conversations; documents distinguish ready, processing, and failed states, with source opening, retry, and explicit removal confirmation.",
            "**Window switching no longer interrupts active replies.** Returning to the desktop window no longer mislabels an active reply as interrupted or leaves subsequent messages queued because of that false state. OpenCode-specific controls remain stable during conversation refreshes, reducing layout shifts.",
            "**Complete scrolling in long conversations.** Long reasoning, tool records, and replies no longer cause the scrollbar to end early because of estimated message heights. All loaded content remains reachable without switching conversations to refresh the layout."
          ],
          "notices": [
            "**Upgrade Desktop before updating the Agent.** Agent `0.11.19` provides Linux x64, Linux arm64, and macOS arm64 packages through one signed catalog. Older Desktop versions cannot parse the new catalog containing macOS entries; upgrade to Desktop `0.12.5` before checking online Agent updates. Existing local packages and historical releases remain available.",
            "Desktop does not embed the Agent or Remote Runtime. Ask remains read-only; Execute uses the selected SSH account’s full permissions.",
            "This release uses the standard six-target desktop matrix and does not include an experimental LoongArch preview.",
            "The repository now includes a standalone interactive ShareServer console prototype. It is not bundled with Desktop or a deployed team-sharing service, and prototype data operations are not persisted."
          ]
        }
      }
    },
    {
      "version": "0.12.4",
      "releasedAt": "2026-09-05",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.12.4 新增会话内 HTML 静态预览，并改善长任务活动、大文件预览、长回复与消息队列、远程任务终态，以及 Linux ARM 上的 OpenCode 后续响应速度。"
          ],
          "features": [
            "**会话内 HTML 静态预览。** Agent 和 Subagent 完成回复后，完整 HTML 文档及 HTML 代码块可直接在会话中预览，并支持查看源码和全屏展示。该能力默认开启，可在“平台功能 / 通用设置”中关闭；预览不会执行脚本、访问网络、提交表单或打开窗口。"
          ],
          "fixes": [
            "**长任务活动与大文件预览更完整。** Runtime 工具和 Subagent 活动现在持久保存在 Main SQLite 中，不再受旧版条目数、详情长度或缓存总量限制；旧数据可能已被截断时会明确提示。本地和远程超过 256 KiB 的 UTF-8 文本文件可按页继续加载，不再直接拒绝预览。",
            "**长回复和消息队列不再意外中断。** 直连模型的流式及非流式长回复、工具参数和累计答案不再因固定字节上限失败。上一次运行失败、被中断或未成功启动后，后续消息会释放并继续调度，不再长期停留在等待队列。",
            "**远程任务终态显示更准确。** 托管 SSH 任务断线恢复后，已持久化的完成或失败状态会覆盖旧的本地流式状态；窗口重新聚焦或恢复可见时也会重新读取会话，避免 Agent 已结束但界面仍显示处理中。",
            "**Linux ARM 上 OpenCode 后续响应更快。** 原生能力清单与后续对话请求现在复用同一个 Runtime，不再先启动并销毁临时 OpenCode Server、再为首条消息重复冷启动。"
          ],
          "notices": [
            "**远程项目仍为技术预览。** Host 需要单独发布的 Agent `0.11.18`；Desktop `0.12.4` 不内置 Agent 或远程 Runtime。Ask 保持只读，Execute 表示授权所选 SSH 账号可用的完整工具、进程、网络和写入权限。",
            "**HTML 预览是静态展示。** 预览保留作者样式，但禁止脚本和外部资源；关闭该功能后继续显示 HTML 源码。",
            "**发布平台。** GoodBuddy 使用 Windows、macOS、Linux 的 `x64`/`arm64` 标准发布矩阵。本次不构建独立的 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.12.4 adds static HTML previews in conversations and improves long-run activity, large-file previews, long replies and message queues, remote terminal state, and OpenCode follow-up speed on Linux ARM."
          ],
          "features": [
            "**Static HTML previews in conversations.** Completed Agent and Subagent replies can render complete HTML documents and HTML code blocks directly in the conversation, with source and full-screen views. The feature is enabled by default and can be disabled under Platform Features / General; previews cannot run scripts, access the network, submit forms, or open windows."
          ],
          "fixes": [
            "**Complete long-run activity and large-file previews.** Runtime tool and Subagent activity is now persisted in Main SQLite without the former item-count, detail-length, or total-cache limits, with a clear warning when legacy data may already have been truncated. Local and remote UTF-8 text files larger than 256 KiB can continue loading page by page instead of being rejected.",
            "**Long replies and message queues no longer stop unexpectedly.** Direct-model streaming and non-streaming replies, tool arguments, and accumulated answers no longer fail at fixed byte limits. After a previous run fails, is interrupted, or never starts, later messages are released and dispatched instead of remaining queued indefinitely.",
            "**Accurate terminal state after remote recovery.** After a managed SSH disconnect, persisted completion or failure now replaces stale local streaming state. Conversations are also refreshed when the window regains focus or visibility, avoiding a Processing state after the Agent has already finished.",
            "**Faster follow-up after switching to OpenCode on Linux ARM.** Native capability discovery and later conversation requests now share one Runtime instead of starting and destroying a temporary OpenCode Server before cold-starting another one for the first message."
          ],
          "notices": [
            "**Remote Projects remain a technical preview.** Hosts require the separately published Agent `0.11.18`; Desktop `0.12.4` does not embed the Agent or Remote Runtime. Ask remains read-only, while Execute authorizes the full tools, processes, network access, and writable paths available to the selected SSH account.",
            "**HTML previews are static.** Author styles are preserved, but scripts and external resources are blocked. Disabling the feature keeps the HTML source visible.",
            "**Release platforms.** GoodBuddy uses the standard Windows, macOS, and Linux `x64`/`arm64` release matrix. This release does not include a separate experimental LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.12.3",
      "releasedAt": "2026-09-04",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.12.3 新增按模型连接定制请求 Header 和 Body，并把内置浏览器扩展到 OpenCode 与 Continue；同时改善多会话并发、长任务、远程 Ask/Execute 和 OpenCode 切换体验。"
          ],
          "features": [
            "**模型请求定制。** 每个 LLM 连接现在可以填写有界 JSON 请求头和顶层请求体，用于兼容自建网关、局域网服务和其他需要额外字段的 Provider。直连模型与 Continue 支持 Header 和 Body，本地 OpenCode 与 DeepSeek Harness 使用各自原生支持的 Header，受管 SSH OpenCode 支持两者；认证、协议和 Runtime 自有字段始终优先。",
            "**Runtime 共享内置浏览器。** 内置浏览器现在可以分配给直连模型、GoodBuddy 管理的 OpenCode 和 Continue，并在 Execute 模式使用。用户与 Agent 共享同一个按会话隔离的浏览器页面；关闭 Agent 浏览器能力后，用户仍可在工作栏中手动导航。"
          ],
          "fixes": [
            "**多会话和长任务不再被人为中止。** 同一项目中的不同 OpenCode 会话现在可以并行运行，同一会话内仍保持顺序。OpenCode、Continue 和受管 SSH 任务不再因 GoodBuddy 固定的工具、活动、模型调用或累计输出数量而取消；本机展示达到容量时会给出提示，但原生任务继续运行。",
            "**OpenCode 状态检查和 Runtime 切换更快。** 被动状态检查现在只验证所选路径和模型凭据，不再启动并立即销毁一个 OpenCode Server，减少较慢设备和 Linux ARM Host 上的切换等待。设置中的显式连接测试、原生能力读取和首次实际请求仍执行完整启动与健康检查，失效的自定义路径也不会被其他候选静默替代。",
            "**模型重试和流式结果更可靠。** 直连模型会在尚未输出内容时，对明确发生在请求发送前的临时连接故障执行可取消重试；可能已经到达 Provider 的请求不会自动重放。Continue 事件流溢出后也不会再把后续增量与最终完整答案重复显示。",
            "**远程 Ask 与 Execute 启动边界更一致。** 两种模式现在都直接启动已签名 Runtime，不再要求 Host 安装 bubblewrap。Ask 同时使用 OpenCode Ask 权限和 Agent 原生读取审批保持只读；Execute 不注入该限制，继续使用所选 SSH 账号的完整权限。",
            "**设置保存和能力提示更准确。** 保存或测试期间设置表单会暂时锁定，避免旧响应覆盖新输入；切换 Embedding 连接也不再改写模型凭据、环境覆盖或并发保存的其他字段。无效 Header 会在保存时提示，浏览器卡片会明确标注仅 Execute 可供 Agent 使用。",
            "**工作栏与终端操作恢复一致。** 终端工具栏粘贴改走受信任的 Main/Preload 剪贴板路径，避免 Electron 拒绝浏览器剪贴板权限后无法粘贴。工作栏页签、关闭按钮和“+”入口恢复圆角样式，终端面板与工具栏继续保持直角。"
          ],
          "notices": [
            "**远程项目仍为技术预览。** Host 需要单独发布的 Agent `0.11.17`；Desktop `0.12.3` 不内置 Agent 或远程 Runtime。Ask 保持只读，Execute 表示授权所选 SSH 账号可用的完整工具、进程、网络和写入权限。",
            "**请求定制不是密钥存储。** 自定义 Header 和 Body 会随连接普通设置保存，请勿在其中填写 API Key 或其他密钥。各 Runtime 只应用其原生支持的字段，界面会标明实际生效范围。",
            "**发布平台。** GoodBuddy 使用 Windows、macOS、Linux 的 `x64`/`arm64` 标准发布矩阵。本次不构建独立的 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.12.3 adds per-connection request headers and body fields and extends the built-in browser to OpenCode and Continue, together with smoother concurrent conversations, long tasks, Remote Ask/Execute, and OpenCode switching."
          ],
          "features": [
            "**Custom model requests.** Each LLM connection can now supply bounded JSON headers and top-level body fields for self-hosted gateways, LAN services, and providers that require additional values. Direct models and Continue support both headers and body fields, local OpenCode and DeepSeek Harness use their native header support, and managed SSH OpenCode supports both; authentication, protocol, and Runtime-owned fields always take precedence.",
            "**Built-in browser shared across Runtimes.** The built-in browser can now be assigned to direct models, GoodBuddy-managed OpenCode, and Continue for use in Execute. The user and Agent share the same conversation-isolated page, while disabling Agent browser access still leaves manual workbar navigation available."
          ],
          "fixes": [
            "**Concurrent conversations and long tasks keep running.** Separate OpenCode conversations in one project can now run in parallel while requests within one conversation stay ordered. OpenCode, Continue, and managed SSH tasks are no longer cancelled by GoodBuddy-defined tool, activity, model-call, or accumulated-output counts; reaching local display capacity shows a notice while native execution continues.",
            "**Faster OpenCode status checks and Runtime switching.** Passive status checks now validate only the selected path and model credential instead of starting and immediately destroying an OpenCode Server, reducing switching delays on slower systems and Linux ARM Hosts. Explicit connection tests, native capability discovery, and first real use still perform full startup and health checks, and an invalid custom path is never silently replaced by another candidate.",
            "**Safer model retries and streaming results.** Direct models perform cancellable retries before any visible output only for temporary connection failures proven to occur before request dispatch; requests that may have reached the provider are never replayed automatically. Continue also avoids showing later stream deltas again with a complete fallback response after an event-stream overflow.",
            "**Consistent Remote Ask and Execute startup boundaries.** Both modes now launch the signed Runtime directly without requiring bubblewrap on the Host. Ask stays read-only through both OpenCode Ask permissions and Agent native-read approval, while Execute applies no such override and retains the selected SSH account's full permissions.",
            "**Reliable settings saves and capability guidance.** Settings are temporarily locked while saving or testing so an older response cannot overwrite newer input, and changing the Embedding connection no longer rewrites model credentials, environment overrides, or unrelated concurrent saves. Invalid headers are rejected when saving, and the browser card now clearly identifies Agent access as Execute-only.",
            "**Restored workbar and terminal behavior.** Terminal toolbar paste now uses the trusted Main/Preload clipboard path instead of a browser clipboard call that Electron could reject. Workbar tabs, close buttons, and the “+” entry use rounded styling again, while the terminal panel and toolbar remain square."
          ],
          "notices": [
            "**Remote Projects remain a technical preview.** Hosts require the separately published Agent `0.11.17`; Desktop `0.12.3` does not embed the Agent or Remote Runtime. Ask remains read-only, while Execute authorizes the full tools, processes, network access, and writable paths available to the selected SSH account.",
            "**Request customization is not credential storage.** Custom headers and body fields are saved as ordinary connection settings, so do not place API keys or other secrets in them. Each Runtime applies only the fields it supports natively, and the interface identifies the effective scope.",
            "**Release platforms.** GoodBuddy uses the standard Windows, macOS, and Linux `x64`/`arm64` release matrix. This release does not include a separate experimental LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.12.2",
      "releasedAt": "2026-09-03",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.12.2 新增多终端助手工作栏、直连模型编程 Agent 和本机工具执行环境，并完善浏览器导航、回复复制、Skills 组织与 DeepSeek Harness 模型选择。"
          ],
          "features": [
            "**助手工作栏与多终端。** 在本机或托管 SSH 项目中开发时，可以从右侧工作栏的“+”同时打开多个终端，分别运行开发服务、测试或其他 Shell 任务。每个终端拥有独立目录、输出和生命周期；关闭页签会结束对应 Shell，应用重启后由用户明确选择是否重连。主侧栏与魔法笔记列表也支持拖拽和键盘调宽，魔法笔记列表可隐藏并记住布局。",
            "**直连模型编程 Agent。** 使用直连模型处理代码时，现在可以直接要求它在 Execute 模式修改文件、运行平台 Shell 并根据测试结果继续修复。模型还可以把一个聚焦的编程步骤委派给单层 Subagent，再将结果汇总回当前回复。整个过程继续使用当前模型、项目和工作模式，并保留取消与活动记录。",
            "**本机工具执行环境。** 在新设备缺少 Node.js 或 Python，或者 Skills、stdio MCP 需要固定解释器时，可以在“能力与工具”中选择 GoodBuddy 托管环境或经过验证的自定义解释器。托管 Node.js 随应用提供，托管 Python 在首次需要时按所选原生地址或 OSS 镜像下载，并显示诊断、安装进度、取消和删除操作。该设置只影响新启动的本机 Runtime 与 stdio MCP，不修改系统环境、普通终端或远程 Host。",
            "**浏览器工作栏导航。** 当 Agent 正在使用内置浏览器调研或操作网页时，可以在工作栏中直接返回、刷新、停止加载或输入新地址，不必结束任务后重新开始。地址栏支持完整 URL、裸域名和带端口的主机地址。用户与 Agent 始终操作同一个按会话隔离的浏览器页面，因此用户导航后的页面会成为 Agent 下一步看到的页面。",
            "**复制 AI 回复。** 需要把回答放入文档、Issue 或其他工具时，可以从已完成回复底部一键复制干净的 Markdown 原文。复制内容不包含推理过程、工具日志或引用元数据，避免手动清理。复制完整会话时也使用同一条经过 Main 校验的剪贴板路径，修复沙箱环境下可能无法写入的问题。",
            "**中文“去 AI 味”审校扩展。** 审阅中文报告、方案或产品文案时，内置审校现在可以发现更多分析套话、模糊归因、角色槽位、工程黑话和绝对化承诺。扫描报告按严重级别和问题分类汇总，便于先处理高影响问题，再逐项修改原文。现有跨平台 Python 使用方式保持不变。"
          ],
          "fixes": [
            "**工具环境选择与诊断更准确。** 自定义 Node.js/Python 入口始终可操作，不再隐式采用首个候选；增加 PATH、venv、Conda 和常见安装位置探测，并移除没有实际证据的依赖占位状态。",
            "**DeepSeek Harness 自动沿用兼容模型。** 没有管理员预置连接时，现在跟随默认或首个兼容模型，无需重复配置。",
            "**Skills 与 MCP 入口更清晰。** 两者统一到“能力与工具”；内置 Skills 从 14 个顶层入口精简为 5 个，产品营销类默认停用，已保存的用户设置保持不变。",
            "**界面和产品文案修正。** 修复主侧栏分隔线未对齐，并将官网、README 与智能心跳中的空泛宣传改为直接说明实际能力和运行边界。",
            "**桌面打包更干净。** 不再显示无需处理的重复依赖引用摘要，也不再把已由 Renderer 打包的 AntV G6 依赖树重复复制到安装包。",
            "**OpenCode 启动路径不再重复验证。** 启动本机 OpenCode 时，GoodBuddy 现在直接使用已选择的绝对路径，不再预先检查文件或运行 `--version`；若路径失效，实际启动会返回错误。托管 Linux ARM Host 激活已登记 Runtime 时也继续依据安装或更新阶段验证的 registry 与 manifest，不重新哈希或检查完整二进制。"
          ],
          "notices": [
            "工具执行环境只作用于本机 Skills 和 stdio MCP，不修改系统环境、普通终端或远程 Host。托管 Python 在首次使用时按需下载。",
            "远程项目仍为默认关闭的技术预览，使用单独发布的 Agent `0.11.15`；Desktop `0.12.2` 不内置 Agent 或远程 Runtime。",
            "本次使用标准 Windows、macOS、Linux `x64`/`arm64` 发布矩阵，不构建 LoongArch 实验预览。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.12.2 adds a multi-terminal assistant workbar, direct-model programming agents, and local tool environments, together with improved browser navigation, response copy, Skills organization, and DeepSeek Harness model selection."
          ],
          "features": [
            "**Assistant workbar and multiple terminals.** While developing in a local or managed SSH project, use the “+” in the right workbar to open separate terminals for a development server, tests, or other Shell tasks. Each terminal has its own directory, output, and lifecycle; closing its tab ends that Shell, and an app restart lets the user explicitly choose whether to reconnect. The main sidebar and Magic Notes list also support pointer and keyboard resizing, while the notes list can be hidden and remembers its layout.",
            "**Direct-model programming agents.** When using a direct model for coding, you can now ask it to edit files, run the platform Shell in Execute mode, and continue fixing the result based on test output. The model can also delegate one focused programming step to a single-level Subagent and synthesize the result into the current response. The process retains the current model, project, and work mode, together with cancellation and activity records.",
            "**Local tool environments.** On a new device without Node.js or Python, or when a Skill or stdio MCP needs a specific interpreter, choose a GoodBuddy-managed environment or a validated custom interpreter under Capabilities and Tools. Managed Node.js ships with the app; managed Python downloads on first use from the selected native or OSS source and shows diagnostics, installation progress, cancellation, and removal. The setting affects only newly started local Runtimes and stdio MCP processes, not the system environment, ordinary terminals, or remote Hosts.",
            "**Browser workbar navigation.** While the Agent researches or operates a page in the built-in browser, you can go back, refresh, stop loading, or enter another address in the workbar without ending and restarting the task. The address field accepts full URLs, bare domains, and hosts with ports. The user and Agent always operate the same conversation-isolated page, so the page reached by user navigation is the page the Agent sees next.",
            "**Copy AI responses.** When moving an answer into documentation, an Issue, or another tool, use the bottom action on a completed response to copy clean Markdown source in one step. The copied text excludes reasoning, tool logs, and citation metadata, avoiding manual cleanup. Full-conversation copy uses the same Main-validated clipboard path and fixes writes that could fail in sandboxed sessions.",
            "**Expanded Chinese de-AI writing review.** When reviewing a Chinese report, proposal, or product copy, the bundled checker now detects more template analysis phrases, unsupported attribution, role-slot prose, engineering jargon, and absolute promises. Its report groups findings by severity and category so users can address high-impact issues first and then revise each location. Existing cross-platform Python usage remains unchanged."
          ],
          "fixes": [
            "**More accurate tool-environment selection and diagnostics.** Custom Node.js/Python choices remain operable and no longer adopt the first candidate implicitly. Discovery now covers PATH, venv, Conda, and common locations, and unsupported dependency placeholders are hidden.",
            "**DeepSeek Harness follows a compatible model automatically.** Without an administrator-provided connection, it now uses the default or first compatible model instead of requiring duplicate configuration.",
            "**Clearer Skills and MCP navigation.** Both now live under Capabilities and Tools. Built-in Skills are reduced from 14 top-level entries to five; product-marketing Skills default to off, while saved user settings remain authoritative.",
            "**Interface and product-copy corrections.** Fixed the misaligned main-sidebar divider and replaced vague promotional wording on the website, in the README, and in Smart Heartbeat with direct descriptions of actual behavior and boundaries.",
            "**Cleaner desktop packaging.** Packaging no longer prints non-actionable duplicate dependency summaries or copies the Renderer-bundled AntV G6 dependency graph into desktop packages.",
            "**OpenCode startup paths are no longer revalidated.** Local OpenCode now uses the selected absolute path directly without checking the file or running `--version` first; an invalid path fails at the actual launch. Registered Runtime activation on managed Linux ARM Hosts also continues to rely on the registry and manifest verified during installation or update without rehashing or rechecking the complete binary."
          ],
          "notices": [
            "Tool environments apply only to local Skills and stdio MCP. They do not modify the system environment, ordinary terminals, or remote Hosts. Managed Python is downloaded on demand when first used.",
            "Remote Projects remain an off-by-default technical preview and use the separately published Agent `0.11.15`; Desktop `0.12.2` does not embed the Agent or Remote Runtime.",
            "This release uses the standard Windows, macOS, and Linux `x64`/`arm64` matrix and does not build an experimental LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.11.14",
      "releasedAt": "2026-08-31",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.11.14 让托管 SSH 任务在桌面退出或网络中断后继续在 Host 上运行并恢复到原会话，同时新增默认保存在本地的有界诊断，以及更清晰、可恢复的桌面反馈。"
          ],
          "features": [
            "**托管 SSH 任务在 Host 上继续运行。** Agent 接受 OpenCode 任务后，不再依赖桌面或 SSH relay 持续在线。重新连接会把结果同步到原会话，不会重发 Prompt、模型调用或工具。项目选择器现按 Host 分组，并显示实时 Agent 连接状态和逐项目恢复进度。",
            "**有界诊断与可选反馈附件。** Desktop 与 Agent 分别轮转保存固定阶段和错误码；Agent 诊断可通过只读 `goodbuddy-agent diagnostics --installation-id <id>` 导出为 JSONL。应用内反馈仅在用户勾选时附加小型桌面诊断摘要，且不会附加远端 Agent 日志。"
          ],
          "fixes": [
            "**Runtime 状态检查不再残留服务。** 应用启动和切换 Runtime 后会立即回收仅用于状态检查的 OpenCode 探测实例；实际执行 Runtime 仍按项目复用。",
            "**Host 版本状态无需重复刷新。** 成功读取的 Agent 与 Runtime 版本会在切换设置分类和重启应用后保留，直到用户刷新、编辑或删除 Host，或开始更新远程环境。",
            "**较慢设备上的 OpenCode 启动更可靠。** Linux ARM 与 QEMU 环境的本地 OpenCode 健康检查现在最多等待 30 秒，减少服务已就绪却误报超时。",
            "**更新控制相互独立。** 关闭自动检查不再禁用更新源切换或手动检查；启动检查和打开下载页失败时会明确提示。",
            "**状态、错误与无障碍反馈更清晰。** 聊天顶部区分 Runtime 连接中、就绪与不可用；Ask/Execute 说明与真实权限一致；最近会话、Host、工作区预览、智能心跳和知识库的加载失败提供明确状态与重试，审批和长任务阶段也会通过辅助技术播报。"
          ],
          "notices": [
            "**远程项目技术预览。** 该能力仍默认关闭。Host 必须安装单独发布的 Agent `0.11.14`（Linux x64/arm64）才能使用新的 Agent-owned 恢复；桌面包不内置 Agent 或远程 Runtime。",
            "**离线执行与恢复边界。** Agent 接受任务后，桌面离线期间仍可能继续模型和工具轮次并产生服务商用量，直到任务完成、恢复连接后被取消，或 Host 侧 Runtime 停止。若 Host、Agent 或 Runtime 重启，内存中的进行中任务无法继续，将报告结果未知且不会自动重放。",
            "**诊断默认仅保存在本地。** 诊断不记录对话、Prompt、凭据、文件内容、路径、环境变量、SSH 参数或 Provider 原始响应；只有用户明确勾选时才将有界桌面摘要附加到反馈，Agent 日志不会自动上传。",
            "**发布平台。** Desktop 使用标准 Windows、macOS、Linux 的 x64/arm64 发布矩阵；本次不构建实验性 LoongArch 预览。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.11.14 keeps managed SSH tasks running on the Host when the Desktop exits or the network disconnects, restores them into the original conversation, and adds bounded local-by-default diagnostics plus clearer, recoverable desktop feedback."
          ],
          "features": [
            "**Managed SSH tasks continue on the Host.** After the Agent accepts an OpenCode task, it no longer depends on the Desktop or SSH relay staying online. Reconnection synchronizes results into the original conversation without replaying the prompt, model calls, or tools. The project selector now groups projects by Host and shows live Agent connectivity and per-project recovery progress.",
            "**Bounded diagnostics and optional feedback attachment.** Desktop and Agent keep separate rotating local records of fixed stages and error codes. Agent diagnostics are available as read-only JSONL through `goodbuddy-agent diagnostics --installation-id <id>`. In-app feedback attaches a small Desktop diagnostic summary only when selected, and never includes remote Agent logs."
          ],
          "fixes": [
            "**Runtime status checks no longer leave services running.** GoodBuddy now releases OpenCode probes used only for status checks after startup and Runtime switching. Execution Runtimes remain reusable per project.",
            "**Host version status no longer needs repeated refreshes.** Successful Agent and Runtime version results persist across settings-category changes and app restarts until the user refreshes, edits or removes the Host, or starts a remote-environment update.",
            "**More reliable OpenCode startup on slower systems.** Local OpenCode health checks now wait up to 30 seconds on Linux ARM and QEMU environments, reducing false timeout reports after the service becomes ready.",
            "**Independent update controls.** Turning off automatic checks no longer disables update-source selection or manual checks. Startup-check and download-page failures are now shown explicitly.",
            "**Clearer status, recovery, and accessibility feedback.** The chat header distinguishes Runtime Connecting, Ready, and Unavailable states; Ask/Execute guidance matches actual permissions; load failures for recent conversations, Hosts, workspace previews, Smart Heartbeat, and Knowledge provide explicit states and retries; approval and long-task stages are also announced to assistive technologies."
          ],
          "notices": [
            "**Remote Projects technical preview.** This capability remains off by default. Hosts require the separately published Agent `0.11.14` for Linux x64 or arm64 to use the new Agent-owned recovery. Desktop packages do not embed the Agent or Remote Runtime.",
            "**Offline execution and recovery boundary.** After the Agent accepts a task, model and tool rounds may continue while the Desktop is offline and may continue provider usage until the task completes, is cancelled after reconnection, or the Host-side Runtime stops. If the Host, Agent, or Runtime restarts, in-memory work cannot continue; it reports an unknown outcome and is not replayed automatically.",
            "**Diagnostics remain local by default.** Diagnostics do not record conversations, prompts, credentials, file contents, paths, environment variables, SSH arguments, or raw provider responses. A bounded Desktop summary is attached to feedback only when explicitly selected, and Agent logs are never uploaded automatically.",
            "**Release platforms.** Desktop uses the standard Windows, macOS, and Linux x64/arm64 matrix. This release does not build an experimental LoongArch preview."
          ]
        }
      }
    },
    {
      "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",
      "notes": {
        "zh-CN": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**远程会话完整保存。** 修复托管 SSH 远程项目的工具结果较长时，会话持久化失败并可能让已完成回复仍显示为生成中的问题。Runtime 已返回的工具输入、输出、错误和摘要现在会完整保存到本地会话，不再受额外的会话字段长度限制。",
            "**魔法笔记列表快捷输入更易发现。** 编辑区现在提示输入 `-`、`1.`、`[ ]` 或 `[x]` 后按空格即可创建项目符号、编号或待办清单，并增加行为覆盖以防快捷输入回退。"
          ],
          "notices": []
        },
        "en-US": {
          "highlights": [],
          "features": [],
          "fixes": [
            "**Complete remote conversation saves.** Fixed conversation persistence failing for managed SSH Remote Projects when tool results were long, which could leave a completed reply displayed as still generating. Tool input, output, errors, and summaries already returned by the Runtime are now saved completely without additional conversation-field length limits.",
            "**Discoverable Magic Notes list shortcuts.** The editor now explains that typing `-`, `1.`, `[ ]`, or `[x]` followed by Space creates a bullet, numbered, or checklist item, with behavior coverage to prevent regressions."
          ],
          "notices": []
        }
      }
    },
    {
      "version": "0.11.10",
      "releasedAt": "2026-08-29",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.11.10 修复正式环境中的 Host 更新失败，并显著缩短 Agent 与 OpenCode Runtime 的安装、恢复和日常启动校验。"
          ],
          "features": [],
          "fixes": [
            "**Host 更新恢复可用。** 修复已发布 OpenCode Runtime 在采用阶段被桌面内置锁重复校验而失败的问题。Host 现在使用签名复合包中的版本锁和身份，并在同一条固定 SSH 连接上完成 Agent 启动、健康检查与 Runtime 激活。",
            "**不再重复扫描远端包体。** 每种获取路径只对复合包执行一次完整 SHA-256；Host installer 在解包时验证全部签名与 payload digest，commit 只检查签名元数据。Main 不再通过 SFTP 逐文件读取 Agent/Runtime，后续健康检查、能力发现和每次提问只使用已登记 identity、manifest 与入口元数据。仍保留 SHA-256，提速来自消除重复读取。",
            "**失败后直接重试。** 未完成操作只保留一个小型 staging 清理记录；下次更新先尽力清理旧暂存，再重新 prepare，不保存或回滚 Agent/Runtime 元数据副本。已健康环境不会因暂存清理失败而回滚，安装错误仍会显示具体的有界原因。"
          ],
          "notices": [
            "**技术预览与版本要求。** 远程项目仍默认关闭且为技术预览。Host 需要安装单独发布的 Agent `0.11.9`，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": [
            "GoodBuddy 0.11.10 fixes Host updates in production and substantially reduces Agent and OpenCode Runtime verification during installation, recovery, and normal startup."
          ],
          "features": [],
          "fixes": [
            "**Host updates work again.** Fixed published OpenCode Runtime adoption failing because it was redundantly checked against the desktop-bundled lock. The Host now uses the signed compound package's lock and identity, then starts and health-checks the Agent and activates Runtime on one pinned SSH connection.",
            "**No repeated remote payload scans.** Each acquisition path performs one full SHA-256 pass over the compound package. The Host installer verifies every signature and payload digest while extracting, and commit checks only signed metadata. Main no longer reads Agent or Runtime file trees over SFTP. Later health, capability, and prompt paths use registered identities, manifests, and entrypoint metadata. SHA-256 remains; the speedup comes from eliminating rereads.",
            "**Simple retry after failures.** An unfinished operation keeps only a small staging-cleanup receipt. The next update makes one best-effort cleanup attempt and performs a fresh prepare without saving or restoring copies of Agent or Runtime metadata. A healthy environment is not rolled back for staging cleanup failures, and bounded installer errors retain their concrete cause."
          ],
          "notices": [
            "**Technical preview and version requirement.** Remote Projects remain disabled by default and in technical preview. Hosts require the separately published Agent `0.11.9` for Linux x64 or arm64. Desktop packages do not embed the Agent or remote Runtime, and this release does not build a LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.11.7",
      "releasedAt": "2026-08-29",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.11.7 让托管 SSH 项目切换与多会话使用更顺畅，重做 Host 远程环境准备，并修复中断后排队消息重复发送。"
          ],
          "features": [
            "**更清楚的 Host 远程环境准备。** “设置 → 平台功能 → 远程项目”的 Host 卡片现在通过一个主按钮完成安装、更新或重装，并可选择“自动”“Host 下载”或“GoodBuddy 传输”。“自动”会在 Host 可直连时由 Host 下载，否则通过 GoodBuddy 传输；已验证的本地或离线导入 Agent 包可以复用，已检查到更高的线上兼容版本时仍使用新版。只有点击主按钮才会开始准备。",
            "**Host 与项目管理。** 可从项目选择器直接打开远程项目设置；即使 Host 不可达或远端目录已不存在，也能删除本地项目记录。删除 Host 会先列出关联项目，并在确认后清理这些项目及其本机会话、任务、成果、记忆、计划和心跳关联；不会连接 Host 或删除远端文件。"
          ],
          "fixes": [
            "**更快、更可靠的托管 SSH 会话。** 切换已有远程项目现在只更新本地选择，同一 Host 上兼容的项目与对话会共用连接。重连后空闲会话不再因 `stale-controller` 无法继续，正在执行的提问仍按原位置恢复或重放。",
            "**排队输入不再重复发送。** 如果应用在消息已写入对话、但队列尚未完成时中断，重启恢复会识别已显示的用户消息并清除对应队列项，不会再次发送同一内容；会话分支也不会继承该队列身份。",
            "**中文下载入口。** 中文官网下载卡片在北京 OSS 索引加载期间保持不可点击，不再短暂暴露 GitHub 回退；仅在索引请求或校验失败后切换到 GitHub。英文下载行为不变。"
          ],
          "notices": [
            "**技术预览与版本要求。** 远程项目仍默认关闭且为技术预览。守护进程端的重连恢复需要 Host 安装单独发布的 Agent `0.11.7`，提供 Linux x64 和 arm64；桌面包不内置 Agent 或远程 Runtime，本次不提供 LoongArch 预览。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.11.7 makes managed SSH project switching and multi-conversation use smoother, overhauls Host remote-environment setup, and prevents queued input from being sent twice after an interruption."
          ],
          "features": [
            "**Clearer Host remote-environment setup.** Under Settings → Platform Features → Remote Projects, each Host card uses one action for install, update, or reinstall, plus Automatic, Host download, and GoodBuddy transfer methods. Automatic downloads on the Host when direct access works; otherwise GoodBuddy transfers it. Verified local or imported Agent packages can be reused, but a known newer compatible online version still wins. Preparation starts only when you select the action.",
            "**Host and project management.** Open a remote project's settings directly from the project switcher, and remove its local record even when the Host is unreachable or the remote folder is gone. Deleting a Host now lists linked projects first and, after confirmation, removes their local conversations, Tasks, results, memories, schedules, and Heartbeat links without connecting to the Host or deleting remote files."
          ],
          "fixes": [
            "**Faster, more resilient managed SSH sessions.** Switching an existing remote project is now local-only, and compatible projects and conversations share one Host connection. Reconnection no longer leaves idle sessions blocked by `stale-controller`, while active prompts preserve exact resume or replay behavior.",
            "**Queued input is no longer sent twice.** If the app is interrupted after a message reaches conversation history but before its queue item completes, startup recovery recognizes the displayed user message and removes the matching queue item instead of sending it again. Conversation branches also discard that queue identity.",
            "**Chinese download entry.** Download cards on the Chinese website now remain disabled while the Beijing OSS index is loading instead of briefly exposing the GitHub fallback. They switch to GitHub only if the OSS request or validation fails; English downloads are unchanged."
          ],
          "notices": [
            "**Technical preview and version requirement.** Remote Projects remain disabled by default and in technical preview. Daemon-side reconnection recovery requires the separately published Agent `0.11.7` for Linux x64 or arm64. Desktop packages do not embed the Agent or remote Runtime, and this release has no LoongArch preview."
          ]
        }
      }
    },
    {
      "version": "0.11.6",
      "releasedAt": "2026-08-27",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.11.6 新增侧栏品牌定制，并修复 Anthropic 输出上限与关键界面提示。"
          ],
          "features": [
            "**侧栏品牌定制。** “设置 → 外观 → 品牌定制”现在可修改左上角品牌名称、中英文副标题和本地 Logo，并提供实时预览、默认 Logo、恢复默认及本机持久化。"
          ],
          "fixes": [
            "**Anthropic 输出上限。** Anthropic 直连模型现在可单独配置单次最大输出，不再错误复用上下文窗口；OpenAI 直连请求和工具调用轮次不增加额外配额。",
            "**输入框与品牌提示。** 恢复聊天和生图输入框中的发送、换行与粘贴快捷键说明，并恢复左上角品牌副标题。"
          ],
          "notices": [
            "品牌定制仅保存在当前设备并只影响左上角品牌区；安装包、更新、Agent 和协议身份仍为 GoodBuddy。",
            "Anthropic 单次最大输出未填写时默认使用 32K Token；实际可用上限仍取决于所选模型和服务商。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.11.6 adds sidebar brand customization while fixing Anthropic output limits and essential interface guidance."
          ],
          "features": [
            "**Sidebar brand customization.** Settings → Appearance → Brand customization now lets you change the top-left brand name, Chinese and English subtitles, and a local logo, with live preview, default-logo, full-reset, and on-device persistence."
          ],
          "fixes": [
            "**Anthropic output limits.** Direct Anthropic connections now have an independent per-response maximum output setting instead of incorrectly reusing the context window. OpenAI direct requests and tool rounds receive no additional quota.",
            "**Composer and brand guidance.** Restored send, newline, and paste shortcuts in chat and image-generation input hints, along with the subtitle beneath the top-left brand name."
          ],
          "notices": [
            "Brand customization is stored only on the current device and affects only the top-left brand area. Installer, update, Agent, and protocol identities remain GoodBuddy.",
            "Anthropic per-response maximum output defaults to 32K tokens when left unset. The selected model and provider may enforce a lower limit."
          ]
        }
      }
    },
    {
      "version": "0.11.5",
      "releasedAt": "2026-08-26",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.11.5 修复 Agent 下载兼容性，显示可用 Agent 更新，并调整新安装的默认模型连接。"
          ],
          "features": [
            "**Agent 更新提示。** “设置 → 平台功能 → 远程项目”现在会显示本地 Agent 版本、在线最新兼容版本以及“有更新 / 已是最新”。打开页面只检查小型签名目录，不会自动下载 Agent 包。"
          ],
          "fixes": [
            "**Agent 公钥目录。** 修复 Windows 发布包使用 CRLF 换行时，合法 Agent 下载被拒绝的问题。JSON 排版不再影响信任，Ed25519 公钥、签名、撤销状态和 SHA-256 校验保持不变。",
            "**默认模型连接。** 新安装改用本机 Ollama 兼容连接；升级只替换未配置凭据的旧默认项，用户显式保存的连接保持不变。",
            "**侧栏菜单稳定性。** 修复魔法笔记未完成待办数字出现或隐藏时左侧菜单行高变化的问题；相邻入口现在保持固定位置。"
          ],
          "notices": [
            "本次正式 Agent 版本为 `0.11.5`，提供 Linux x64 与 arm64；大型包继续使用流式 SHA-256 校验，不会降级为 MD5。",
            "`0.11.5` 将另外提供独立的实验性 LoongArch 桌面预览包。它不进入标准更新索引，且不包含 Agent、Remote Runtime、OpenCode 和原生 Canvas。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.11.5 fixes Agent download compatibility, shows available Agent updates, and updates the default model connection for new installations."
          ],
          "features": [
            "**Agent update status.** Settings → Platform Features → Remote Projects now shows the local Agent version, latest compatible online version, and Update available / Up to date status. Opening the page checks only the small signed catalog and never downloads an Agent package automatically."
          ],
          "fixes": [
            "**Agent public-key registry.** Fixed valid Agent downloads being rejected when Windows release packages used CRLF line endings. JSON formatting no longer affects trust, while Ed25519 keys, signatures, revocations, and SHA-256 verification remain unchanged.",
            "**Default model connection.** New installations now use a local Ollama-compatible connection. Upgrades replace only the old default when it has no configured credential, while explicitly saved user connections remain unchanged.",
            "**Stable sidebar navigation.** Fixed the Magic Notes incomplete-todo count changing the sidebar row height when the badge appeared or disappeared. Neighboring navigation items now remain in place."
          ],
          "notices": [
            "The formal Agent version is `0.11.5`, with Linux x64 and arm64 packages. Large packages continue to use streaming SHA-256 and are not downgraded to MD5.",
            "A separate experimental LoongArch desktop preview will be provided for `0.11.5`. It remains outside the standard update index and excludes Agent, Remote Runtime, OpenCode, and native Canvas."
          ]
        }
      }
    },
    {
      "version": "0.11.4",
      "releasedAt": "2026-08-26",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.11.4 新增托管 SSH 远程项目、内置 Granite 向量模型、会话分支和应用内反馈，并改进魔法笔记、知识图谱、模型下载及桌面布局。"
          ],
          "features": [
            "**托管 SSH 远程项目（技术预览）。** 在“设置 → 平台功能”中启用后，可管理 Linux x64/arm64 Host，并通过 Ask 或 Execute 使用签名 Agent 与兼容的 OpenCode Runtime。Agent 包需在设置中手动下载、更新或离线导入，打开项目不会自动下载；SSH 断线后，Host 上的任务可继续运行并恢复未同步输出。",
            "**内置 Granite 向量模型。** 可下载、校验、离线导入、移除并运行内置 Granite 多语言 Embedding 模型，同时继续支持自定义 OpenAI 兼容向量连接和检索诊断。",
            "**独立会话分支。** 本地会话可通过“在新会话中继续”创建独立分支并保留来源，不复制原会话的任务、队列、运行记录、记忆或成果归属。",
            "**应用内反馈。** 可在“关于与更新”中提交问题、建议或体验反馈；邮箱和单张截图均为可选项，失败后会保留草稿供重试。",
            "**魔法笔记待办数量。** 左侧导航可显示未完成待办数量，并可在“平台功能 → 魔法笔记”中关闭；超过 99 项时显示“99+”。",
            "**DeepSeek Harness Runtime。** 内置 DeepSeek Harness 已从 `0.1.0-rc.6` 升级至 `0.1.0-rc.8`；GoodBuddy 同步适配新版图片附件限制，现有受控 Host、Ask/Execute 与插件边界保持不变。",
            "**Linux 与 LoongArch 构建。** Linux x64/arm64 桌面版新增 RPM 安装包，并提供独立、可复现的 Debian loong64 实验性预览构建流程。"
          ],
          "fixes": [
            "**向量模型下载反馈。** 修复点击内置向量模型“下载”后界面看似无响应的问题；现在会立即显示下载来源、模型信息、实时进度和取消操作，行为与语音模型下载一致。",
            "**长文档知识图谱。** 图谱抽取改为按 Token 分批、重试和超时处理；图谱失败不再导致新文档全文检索一并失败，并改进中文检索与关系上下文。",
            "**ModelScope 语音模型。** 修复选择 ModelScope 后部分语音模型仍无法下载的问题，六个支持的模型现在均使用固定且经过完整性验证的来源。",
            "**魔法笔记与平台功能。** 待办勾选会立即同步源笔记，同时修复暗色模式输入提示过暗，以及“平台功能”页签可能不可见的问题。",
            "**助手工作栏与主导航。** 工作栏保持停靠和可调整大小，统一切换与任务入口，恢复左侧全高导航，并避免遮挡聊天输入区。",
            "**界面信息与会话列表。** 恢复关键页面说明和视觉层级；会话更多操作仅在悬停、键盘聚焦或菜单打开时显示，为标题和时间保留更多空间。"
          ],
          "notices": [
            "远程项目默认关闭且完全可选。首次使用前必须手动准备兼容的 `.gbagent` 包，下载源跟随“关于与更新”中的 GitHub 或镜像节点选择；本次正式 Agent 版本为 `0.11.4`。",
            "LoongArch 仍是独立实验性预览，不进入标准更新索引；当前不包含 Agent、Remote Runtime、OpenCode 和原生 Canvas。本次 `0.11.4` 不提供 LoongArch 预览包。",
            "反馈不会自动附带会话、日志、文件、设置或凭据。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.11.4 adds managed SSH remote projects, a built-in Granite embedding model, conversation branching, and in-app feedback, with improvements to Magic Notes, knowledge graphs, model downloads, and desktop layout."
          ],
          "features": [
            "**Managed SSH remote projects (technical preview).** Enable Remote Projects under Settings → Platform Features to manage Linux x64/arm64 Hosts and use signed Agents with a compatible OpenCode Runtime in Ask or Execute. Agent packages must be downloaded, updated, or imported manually in Settings; opening a project never downloads one automatically. Host work can continue through an SSH disconnect and recover unsynchronized output.",
            "**Built-in Granite embedding model.** Download, verify, import offline, remove, and run the managed Granite multilingual embedding model while retaining custom OpenAI-compatible vector connections and retrieval diagnostics.",
            "**Independent conversation branches.** Continue a local conversation in a new branch that records its source without copying Tasks, queued input, runs, memories, or artifact ownership.",
            "**In-app feedback.** Submit problems, ideas, or experience notes from About & Updates. An email address and one screenshot are optional, and failed submissions retain the draft for retry.",
            "**Magic Notes todo count.** The primary navigation can show incomplete todo counts, with an option under Platform Features → Magic Notes to hide the badge. Counts above 99 display as “99+”.",
            "**DeepSeek Harness Runtime.** The bundled DeepSeek Harness is updated from `0.1.0-rc.6` to `0.1.0-rc.8`. GoodBuddy now matches the updated image attachment limits while preserving its controlled Host, Ask/Execute, and plugin boundaries.",
            "**Linux and LoongArch builds.** Linux x64/arm64 desktop releases add RPM installers, and the repository now provides a separate reproducible experimental Debian loong64 preview build."
          ],
          "fixes": [
            "**Embedding model download feedback.** Fixed the built-in model Download action appearing unresponsive. It now immediately shows the source, model details, live progress, and cancellation, matching speech model downloads.",
            "**Long-document knowledge graphs.** Graph extraction now uses token-bounded batches, retries, and timeouts. Graph failures no longer discard a new document’s full-text index, and Chinese retrieval and relationship context are improved.",
            "**ModelScope speech models.** Fixed supported speech models remaining unavailable after selecting ModelScope. All six models now use fixed, integrity-verified sources.",
            "**Magic Notes and Platform Features.** Todo changes immediately refresh their source note. Dark-mode placeholder contrast and disappearing Platform Features tabs are also fixed.",
            "**Assistant workbar and primary navigation.** The workbar remains docked and resizable, uses consistent toggle and Task entry points, restores the full-height sidebar, and no longer crowds the composer.",
            "**Interface guidance and conversation rows.** Restored essential workflow guidance and visual hierarchy. Conversation actions now appear only on hover, keyboard focus, or while their menu is open, leaving more room for titles and timestamps."
          ],
          "notices": [
            "Remote Projects are disabled by default and remain optional. A compatible `.gbagent` package must be prepared manually before first use, and its source follows the GitHub or mirror selection in About & Updates. This release uses the formal Agent version `0.11.4`.",
            "LoongArch remains a separate experimental preview outside the standard update index. It currently excludes Agent, Remote Runtime, OpenCode, and native Canvas. No LoongArch preview package is provided for `0.11.4`.",
            "Feedback never automatically includes conversations, logs, files, settings, or credentials."
          ]
        }
      }
    },
    {
      "version": "0.11.0",
      "releasedAt": "2026-08-20",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.11.0 新增与会话关联的定制计划任务和持久发送队列，扩展智能心跳范围与本地模型下载源，并全面提升桌面工作流的可靠性。"
          ],
          "features": [
            "**定制计划任务。** 可将单次、每日或每周任务关联到当前或新会话；重复触发会复用同一 Task，将文本结果写回会话，并可从会话列表和 Task Center 统一查看与管理。",
            "**会话发送队列。** 回复生成期间仍可继续发送消息；普通消息与到期计划任务会按 Conversation 持久排队，支持删除待发送项，或中断当前回复并优先执行指定项，重启后仍可恢复。",
            "**智能心跳范围。** 心跳计划现在从唯一的权威入口选择 Global 或一个、多个 Project；旧配置和历史会无损迁移，项目级记忆与行动输出会严格写入所选范围。",
            "**本地模型下载源。** 可在“设置 → 平台功能 → 通用设置”中选择 ModelScope 或 Hugging Face，统一用于后续语音输入与 OCR 模型下载；下载不会静默换源或混合不同来源的文件。",
            "**可配置全局快捷键。** 可启停、录制或恢复 GoodBuddy 的全局唤起快捷键；冲突或保存失败时会保留上一组可用快捷键。",
            "**更清楚的项目切换。** 项目选择器现在按本地项目和远程通道分组，并显示目录或通道来源，切换时更容易辨认目标。"
          ],
          "fixes": [
            "**真实模型验证与凭据保留。** “保存并测试”现在执行有界的真实文本或图片生成并校验输出，不再只把接口成功响应视为模型可用；修改地址或临时关闭认证时，加密 API Key 会继续随连接保留，直到用户明确清除或删除连接。",
            "**多专家结果展示。** 并行分析现在可以展开查看每位专家的完整输出，并在其下方显示和保存总 Agent 的综合结果。",
            "**Runtime 与数据可靠性。** 强化 Runtime 子进程退出、配置原子回滚、模型包安装及持久化写入恢复；设置、知识库、魔法笔记和智能心跳失败时会更可靠地保留状态与未保存草稿。",
            "**成果与通道回复。** 普通本地和消息通道回复不再重复出现在成果栏，已有重复项只会隐藏而不会删除；完整通道结果也不再被公共服务统一截断为 4,000 字符。",
            "**内置浏览器设置。** 内置浏览器已移至直连模型工具，并明确它只操作 GoodBuddy 的隔离浏览器；暂时隐藏尚未生效的托管浏览器配置，同时保留原有开关状态和底层数据。",
            "**界面与加载体验。** 助手工作栏可在宽屏上使用更多空间，打开会话时会可靠保持在底部；同时改进对话框、键盘操作、无障碍状态和重型页面的按需加载。"
          ],
          "notices": [
            "**计划任务限制。** 当前支持单次、每日和每周计划，应用关闭期间不能执行任务；新建任务默认使用 Execute，但仍遵守所选 Runtime、工具授权和高风险审批边界。",
            "**模型测试可能产生用量。** 真实文本或图片测试会调用所选模型服务，可能产生少量 Token 或图片生成费用。",
            "**模型下载源默认值。** 升级后默认使用 ModelScope；切换来源只影响之后启动的语音与 OCR 下载，已安装模型、正在进行的下载和 ZIP 导入不受影响。",
            "**全局快捷键兼容性。** 默认组合为 `CommandOrControl+Shift+Space`；若操作系统或桌面环境不支持注册，仍可通过窗口或托盘使用 GoodBuddy。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.11.0 adds conversation-backed custom scheduled tasks and a durable input queue, expands Smart Heartbeat scoping and local model download sources, and strengthens reliability across desktop workflows."
          ],
          "features": [
            "**Conversation-backed scheduled tasks.** Create once, daily, or weekly custom tasks for the current or a new conversation. Recurring runs reuse one Task, write text results back to the conversation, and can be viewed and managed from the conversation list and Task Center.",
            "**Durable conversation queue.** Send more messages while a reply is running. Messages and due Scheduled Tasks are persisted and serialized per Conversation, can be removed or promoted by interrupting the current reply, and recover after restart.",
            "**Scoped Smart Heartbeat.** Heartbeat plans now select Global or one or more Projects from a single authoritative editor. Existing configurations and history migrate without data loss, while project-level memory and actions remain within the selected scope.",
            "**Local model download sources.** Settings → Platform Features → General now lets you choose ModelScope or Hugging Face for subsequent speech-input and OCR model downloads. Downloads never silently switch sources or mix files from different sources.",
            "**Configurable global shortcut.** Enable, disable, record, or restore GoodBuddy’s global activation shortcut. If registration conflicts or saving fails, the last working shortcut remains active.",
            "**Clearer project switching.** The project selector now groups local projects and remote channels and shows folder or channel context, making the intended destination easier to identify."
          ],
          "fixes": [
            "**Real model verification and retained credentials.** Save & Test now performs bounded real text or image generation and validates the output instead of treating a successful endpoint response as proof that the model works. Encrypted API keys remain with their connection across URL or authentication changes until explicitly cleared or the connection is deleted.",
            "**Visible expert results.** Parallel analyses now let you expand each expert’s complete output, with the lead Agent’s synthesis displayed beneath them and saved with the conversation.",
            "**Runtime and data reliability.** Improved Runtime child-process shutdown, atomic configuration rollback, model-package installation, and persisted-write recovery. Settings, Knowledge, Magic Notes, and Smart Heartbeat now preserve state and unsaved drafts more reliably when an operation fails.",
            "**Results and channel replies.** Ordinary local and channel replies no longer appear again in Results, while legacy duplicates are hidden without deleting data. Complete channel output also reaches each platform adapter instead of being truncated by the shared service at 4,000 characters.",
            "**Built-in browser settings.** The built-in browser now appears under direct-model tools and is clearly identified as GoodBuddy’s isolated browser. Inactive managed-profile controls are hidden while preserving the existing switch state and stored data.",
            "**Layout and loading experience.** The assistant workbar can use more space on wide screens, opened conversations reliably remain at the bottom, and dialogs, keyboard behavior, accessibility states, and lazy loading for heavy pages have been improved."
          ],
          "notices": [
            "**Scheduled-task limits.** Current schedules support once, daily, and weekly triggers and cannot run while the app is closed. New tasks default to Execute but continue to honor the selected Runtime, tool authorization, and high-risk approval boundaries.",
            "**Model tests may incur usage.** Real text or image tests call the selected model provider and may consume a small number of tokens or incur image-generation charges.",
            "**Default model download source.** Upgrades default to ModelScope. Changing the source affects only subsequent speech and OCR downloads; installed models, active downloads, and ZIP imports are unchanged.",
            "**Global shortcut compatibility.** The default accelerator is `CommandOrControl+Shift+Space`. If the operating system or desktop environment cannot register it, GoodBuddy remains available from its window or tray."
          ]
        }
      }
    },
    {
      "version": "0.10.4",
      "releasedAt": "2026-08-17",
      "notes": {
        "zh-CN": {
          "highlights": [
            "GoodBuddy 0.10.4 新增可选的镜像节点，并改进最近对话的时间显示。"
          ],
          "features": [
            "**镜像节点与更新源。** 在“关于与更新”中可选择 GitHub（默认）或镜像节点。手动检查、启动时检查和打开下载页会使用同一选择，应用仍只检查版本，不会自动下载安装。"
          ],
          "fixes": [
            "**最近对话时间。** 侧栏中的当天对话显示本地时间，较早对话显示日期，跨年记录同时显示年份；悬停可查看完整日期和时间。"
          ],
          "notices": [
            "**默认更新源。** 现有用户升级后仍默认使用 GitHub；只有手动选择后才会使用镜像节点。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.10.4 adds an optional mirror node and improves how recent conversation times are displayed."
          ],
          "features": [
            "**Mirror node and update source.** About & Updates now lets you choose GitHub (default) or the mirror node. Manual checks, startup checks, and the download page use the same selection. GoodBuddy still checks versions only and never downloads or installs updates automatically."
          ],
          "fixes": [
            "**Recent conversation times.** Today’s conversations show local time in the sidebar, older conversations show the date, and entries from another year include the year. Hovering reveals the complete date and time."
          ],
          "notices": [
            "**Default update source.** Existing users continue to use GitHub after upgrading. The mirror node is used only after it is selected manually."
          ]
        }
      }
    },
    {
      "version": "0.10.1",
      "releasedAt": "2026-08-17",
      "notes": {
        "zh-CN": {
          "highlights": [
            "0.10.1 重点完善多 Runtime 工作流。OpenCode、Continue 和 DeepSeek Harness 的能力、MCP、插件与上下文管理现在可以在 GoodBuddy 内统一查看和配置，长对话、多会话和任务通知也更加连贯。"
          ],
          "features": [
            "**Runtime 能力概览。** 切换 Runtime 或排查不可用工具，设置页会展示内置 OpenCode、Continue 和 DeepSeek Harness 实际提供的 Agents、Tools、Commands、Rules、Prompts、Skills、MCP 等能力，并提供 Runtime 默认项与上下文压缩配置。",
            "**DSH 插件管理。** DeepSeek Harness 的工具能力可以通过默认关闭的插件市场扩展，支持搜索、安装、更新和启停 npm 插件，并填写插件所需的 JSON 配置。",
            "**按 Runtime 分配 MCP。** 知识库、魔法笔记等内置 MCP 可以分别分配给直连模型、OpenCode 和 Continue；自定义 MCP 也能分配给 OpenCode、Continue 和 DeepSeek Harness，同一套工具服务不必重复配置。",
            "**上下文压缩。** 内置 OpenCode 默认自动整理上下文，也可手动触发；Continue 可手动生成并复用 GoodBuddy 摘要；直连模型是否自动压缩由用户决定。界面会分别显示本次调用用量和压缩后的对话估算。",
            "**DSH 图片输入。** 支持图片的模型连接可以接收经过校验的 JPEG/PNG 截图与图片；文本模型会在调用前明确提示不支持图片，避免提交后才发现无法处理。",
            "**多会话并行处理。** 多个会话可以在后台运行任务，另一个会话仍可继续聊天。侧栏会标记活动和未读状态，聊天内容与页面位置也会按会话保留。",
            "**Runtime 用量统计。** 运行记录会按 Runtime 和模型归类每次调用，连续工具调用与上下文摘要不再合并计数，并会显示归一化的缓存命中率。"
          ],
          "fixes": [
            "**Windows 通知激活。** 任务完成后，点击系统通知会回到 GoodBuddy，不再打开 Electron 默认页；开发版、解包版和安装版也使用相互隔离的通知身份。",
            "**直连模型连续对话。** 较长的连续对话和消息重编辑不再因本地消息 ID 被发送给模型服务而失败，工具调用与上下文摘要也不会覆盖前序用量记录。",
            "**DSH 插件与 MCP 稳定性。** 多个 DSH 插件共同运行、MCP 返回大量分页工具，启动、取消和退出依然可靠；单个插件失败会被隔离，循环游标和未释放会话也会得到处理。",
            "**设置界面一致性。** DSH 搜索文字不再被图标遮挡，正常状态不会重复显示说明，导航高亮与页面边距也保持一致。",
            "**跨架构发布包。** Windows、macOS 和 Linux 的 x64、arm64 安装包会校验 Canvas、Koffi 等原生依赖，避免构建成功后加载错误架构的二进制文件。",
            "**魔法笔记编辑体验。** 更高且可纵向拖动的编辑区域为长笔记留出空间，数字字号选项、工具栏和分栏间距也更加清楚；AI 评论区会在笔记详情就绪前明确显示加载状态。"
          ],
          "notices": [
            "**DSH 插件市场。** 该功能仍处于预览阶段并默认关闭。第三方插件的安装脚本、初始化代码和 Execute 工具会以当前用户权限运行，请只安装可信的包。关闭市场不会自动停用或卸载已有插件。",
            "**工作模式权限。** Ask 模式仍保持只读；Execute 模式下，经过批准的工具会以当前用户权限操作文件、运行命令或访问外部服务。",
            "**上下文压缩默认行为。** 内置 OpenCode 默认启用原生自动压缩；直连模型的自动压缩默认关闭；Continue 当前不会自动生成摘要，需要手动执行压缩。压缩可能调用所选模型并产生额外 Token 用量，GoodBuddy 中的原始聊天记录不会删除。",
            "**Runtime 兼容性。** 外部 OpenCode Server 当前只提供连接状态，Continue 暂不支持静态发现原生 Tools；DeepSeek Harness 暂不支持内置 MCP，但可在 Execute 模式使用已分配的自定义 MCP。"
          ]
        },
        "en-US": {
          "highlights": [
            "GoodBuddy 0.10.1 strengthens multi-Runtime workflows. OpenCode, Continue, and DeepSeek Harness capabilities, MCP, plugins, and context controls can now be viewed and configured in one place, with smoother long conversations, parallel work, and task notifications."
          ],
          "features": [
            "**Runtime capability overview.** Switching Runtimes or diagnosing unavailable tools now reveals the Agents, Tools, Commands, Rules, Prompts, Skills, MCP, and other capabilities actually provided by managed OpenCode, Continue, and DeepSeek Harness, along with Runtime defaults and context controls.",
            "**DSH plugin management.** DeepSeek Harness can be extended through the default-off plugin marketplace, with npm plugin search, installation, updates, enablement, and JSON configuration.",
            "**MCP assignment by Runtime.** Built-in MCP servers such as Knowledge and Magic Notes can be assigned individually to direct models, OpenCode, and Continue. Custom MCP can also be shared with OpenCode, Continue, and DeepSeek Harness without duplicating service configuration.",
            "**Context compaction.** Managed OpenCode compacts context automatically by default and also provides a manual action. Continue can manually create and reuse GoodBuddy summaries, while users decide whether to enable automatic compaction for direct models. Latest-call usage and compressed-conversation estimates are shown separately.",
            "**DSH image input.** Image-capable model connections accept validated JPEG/PNG screenshots and images. GoodBuddy rejects image input with a clear message before invoking a text-only model.",
            "**Parallel conversations.** Multiple conversations can keep running tasks in the background while you continue chatting in another. The sidebar marks active and unread conversations, while chat content and page position remain preserved per conversation.",
            "**Runtime usage reporting.** Run History groups every call by Runtime and model. Consecutive tool calls and context summaries remain separate usage records, with normalized prompt-cache hit rates for easier comparison."
          ],
          "fixes": [
            "**Windows notification activation.** Clicking a task-completion notification now opens GoodBuddy instead of Electron’s default page. Development, unpacked, and installed builds also use isolated notification identities.",
            "**Direct-model follow-ups.** Long conversations and edited messages no longer fail because local message IDs reached model providers. Tool and context-summary calls also preserve earlier usage records.",
            "**DSH plugin and MCP reliability.** Startup, cancellation, and shutdown remain reliable with several DSH plugins or large paginated MCP tool sets. Plugin failures are isolated, cursor loops are bounded, and sessions are released.",
            "**Settings consistency.** DSH plugin search text no longer overlaps its icon, healthy status descriptions are no longer duplicated, and navigation highlights and page gutters remain aligned.",
            "**Cross-architecture packages.** Windows, macOS, and Linux packages validate Canvas, Koffi, and related native dependencies for x64 and arm64, preventing successful builds from loading binaries for the wrong architecture.",
            "**Magic Notes editing.** The editor is taller and vertically resizable, with numeric font-size choices and clearer toolbar and pane spacing. The AI comments pane now shows an explicit loading state until note details are ready."
          ],
          "notices": [
            "**DSH plugin marketplace.** This preview feature is disabled by default. Third-party install scripts, initialization code, and Execute tools run with current-user permissions, so install only trusted packages. Disabling the marketplace does not disable or remove installed plugins.",
            "**Work mode permissions.** Ask remains read-only. In Execute, approved tools can modify files, run commands, or access external services with current-user permissions.",
            "**Context compaction defaults.** Managed OpenCode enables native automatic compaction by default. Automatic compaction for direct models is disabled by default, while Continue requires manual compaction to create a summary. Compaction may call the selected model and incur additional token usage; original chat history in GoodBuddy is not deleted.",
            "**Runtime compatibility.** External OpenCode Servers currently expose connection status only, and Continue cannot statically discover native Tools. DeepSeek Harness does not yet support built-in MCP, but assigned custom MCP is available in Execute mode."
          ]
        }
      }
    },
    {
      "version": "0.9.3",
      "releasedAt": "2026-08-15",
      "notes": {
        "zh-CN": {
          "features": [
            "新增直连模型上下文控制：可在达到阈值时自动摘要较早对话，保留最近完整问答，并可为每个模型配置上下文上限及选择摘要模型；原始聊天记录不会删除。",
            "重新设计“运行记录”，按项目、任务和会话组织执行详情，新增共享时间轴、活动筛选、节点详情，以及按项目、会话和模型汇总的 Token 用量视图。",
            "统一通道项目设置入口；微信 ClawBot、企业微信和钉钉项目的名称、说明、Runtime 与工作模式现在会在项目设置和通道设置之间保持同步。",
            "调整 Agent Runtime 执行方式：DeepSeek Harness、OpenCode 和 Continue 在 Execute 模式下通过现有授权控制，以当前用户权限运行工具。"
          ],
          "fixes": [
            "提升大型会话与流式回复的响应速度：增量保存会话、限制首屏渲染量、按帧合并更新、并行启动前置任务，并在空闲时预载页面。",
            "修复 OpenAI Responses 与 Anthropic 的工具轮次等待完整响应的问题；文本和推理现在会在工具执行及后续轮次中持续流式显示。",
            "修复打包应用启动时缺少 DeepSeek Harness bundle manifest、导致主进程无法加载的问题。",
            "强化流式事件与会话保存顺序：首段内容更快显示，结束状态会可靠持久化，并降低退出应用或长会话期间状态丢失的风险。"
          ]
        },
        "en-US": {
          "features": [
            "Added direct-model context controls that summarize earlier conversation at a configurable threshold, preserve recent full turns, support per-model context limits, and allow a dedicated summary model without deleting chat history.",
            "Redesigned Run History around projects, tasks, and conversations, with a shared activity timeline, filters, node details, and token-usage views grouped by project, conversation, or model.",
            "Unified channel project settings so WeChat ClawBot, WeCom, and DingTalk project names, descriptions, Runtime selections, and work modes stay synchronized across both settings entry points.",
            "Updated Agent Runtime execution so DeepSeek Harness, OpenCode, and Continue run tools with current-user permissions in Execute mode under the existing authorization controls."
          ],
          "fixes": [
            "Improved responsiveness for large conversations and streamed replies with incremental persistence, bounded initial rendering, frame-paced updates, parallel startup prerequisites, and idle route preloading.",
            "Fixed OpenAI Responses and Anthropic tool rounds waiting for a complete response; text and reasoning now stream continuously through tool execution and continuation rounds.",
            "Fixed packaged-app startup failures caused by a missing DeepSeek Harness bundle manifest.",
            "Strengthened streaming-event and conversation-save ordering so initial content appears sooner, terminal state is persisted reliably, and state is less likely to be lost during quit or long conversations."
          ]
        }
      }
    },
    {
      "version": "0.9.2",
      "releasedAt": "2026-08-14",
      "notes": {
        "zh-CN": {
          "features": [
            "新增长对话“到底部”浮动按钮；阅读较早消息时，流式回复会保持当前位置，只有停留在底部附近时才自动跟随最新内容，并遵循系统的减少动态效果偏好。"
          ],
          "fixes": [
            "精简 DeepSeek Harness 设置说明，移除与连接配置重复的兼容性提示。",
            "修复应用内 0.9.0 与 0.9.1 更新说明内容重复的问题。"
          ]
        },
        "en-US": {
          "features": [
            "Added a floating “Scroll to bottom” control for long conversations; streamed responses now preserve the reader’s position unless they remain near the bottom, and the control respects the system reduced-motion preference."
          ],
          "fixes": [
            "Simplified the DeepSeek Harness settings by removing a compatibility notice that duplicated the connection guidance.",
            "Fixed duplicate in-app release-note content between versions 0.9.0 and 0.9.1."
          ]
        }
      }
    },
    {
      "version": "0.9.1",
      "releasedAt": "2026-08-14",
      "notes": {
        "zh-CN": {
          "features": [
            "新增 DeepSeek Harness 开发者预览 Runtime，支持安全的 OpenAI 兼容 Chat Completions 连接、任意模型名称、Skills 与自定义 MCP，并保留 Ask 只读、Execute 授权、取消和工作区沙箱边界。",
            "新增自然语言配置工具，助手可查看脱敏设置并规划应用偏好、Skill 与 MCP 变更；配置只在原生审批通过后应用，凭据不会暴露给模型。",
            "GoodBuddy 原创源代码现以 0BSD 许可证开放，并补充英文项目总览以及自动化、会话监督、分区记忆、并行实验和持续学习等产品规划文档。"
          ],
          "fixes": [
            "修复模型工具调用期间流式推理内容可能折叠或不可见的问题，并让推理区域在生成时自动跟随最新内容。",
            "修复从通道入口打开设置时未定位到所选企业微信、钉钉或微信页面的问题，并更正微信二维码扫码提示。",
            "优化简体中文界面的系统字体、字号和行高，改善 Windows 与 macOS 上的小字号可读性和排版一致性。",
            "修复 Windows arm64 发布构建缺少目标架构原生依赖、导致该平台安装包无法生成的问题。"
          ]
        },
        "en-US": {
          "features": [
            "Added the DeepSeek Harness preview Runtime with secure OpenAI-compatible Chat Completions, arbitrary model names, Skills, and custom MCP, while preserving read-only Ask, authorized Execute, cancellation, and workspace sandbox boundaries.",
            "Added natural-language configuration tools that let the assistant inspect sanitized settings and plan changes to app preferences, Skills, and MCP; changes apply only after native approval without exposing credentials to models.",
            "Released GoodBuddy's original source under the 0BSD License and expanded the English overview and product plans for automation, conversation supervision, partitioned memory, parallel experiments, and continuous learning."
          ],
          "fixes": [
            "Fixed streamed reasoning becoming hidden during model tool calls, and kept the reasoning panel following the latest content while generation is in progress.",
            "Fixed channel shortcuts opening the wrong settings page for WeCom, DingTalk, or WeChat, and corrected the WeChat QR-code scan guidance.",
            "Improved Simplified Chinese typography with platform-native UI fonts, refined sizes, and line heights for clearer, more consistent text on Windows and macOS.",
            "Fixed missing target-architecture native dependencies in Windows arm64 release builds, which prevented installers for that platform from being produced."
          ]
        }
      }
    },
    {
      "version": "0.8.20",
      "releasedAt": "2026-08-13",
      "notes": {
        "zh-CN": {
          "features": [
            "全面升级本地知识库，新增中文、全文、向量与知识图谱混合检索、本地及学习型重排、检索诊断工作台，并重新组织文档、图谱、任务和索引工作区。",
            "重新组织 MCP 设置，并支持为自定义 MCP 服务选择启用动态工具列表更新；现有服务默认保持原有行为。",
            "新增 KaTeX 数学公式渲染，支持在聊天 Markdown 中显示行内公式和块级公式。",
            "新增交互式 Mermaid 图表渲染，支持查看源码、放大、缩放和拖动，并在渲染失败时回退到源码。"
          ],
          "fixes": [
            "修复异常退出后会话、笔记、运行中消息、工具调用和定时任务状态可能丢失或不一致的问题。",
            "修复企业微信、钉钉和微信等远程通道消息发送失败后可能丢失的问题，未投递消息现在会持久化并重试。",
            "修复知识索引任务在重启后状态不准确，以及索引重建中断或失败时可能暴露不完整结果的问题。",
            "提升模型流式响应和文档提取的稳定性，对异常大的响应、工具参数和文档提供明确限制及错误提示。",
            "修复直连模型使用工具时推理内容流式显示不完整、聊天宽表格溢出，以及部分设置和作用域工具保存不可靠的问题。"
          ]
        },
        "en-US": {
          "features": [
            "Upgraded the local knowledge base with hybrid Chinese, full-text, vector, and knowledge-graph retrieval, local and learning-based reranking, a retrieval diagnostics workbench, and reorganized document, graph, task, and indexing workspaces.",
            "Reorganized MCP settings and added opt-in dynamic tool-list updates for custom MCP services, while preserving existing behavior by default.",
            "Added KaTeX math rendering for inline and block formulas in chat Markdown.",
            "Added interactive Mermaid diagram rendering with source viewing, zooming, panning, and source fallback when rendering fails."
          ],
          "fixes": [
            "Fixed lost or inconsistent conversation, note, in-progress message, tool-call, and scheduled-task states after an unexpected shutdown.",
            "Fixed messages being lost after delivery failures on remote channels such as WeCom, DingTalk, and WeChat; undelivered messages are now persisted and retried.",
            "Fixed inaccurate knowledge-index task states after restart and incomplete results becoming visible when an index rebuild was interrupted or failed.",
            "Improved stability for model streaming and document extraction by enforcing clear limits and errors for unusually large responses, tool arguments, and documents.",
            "Fixed incomplete streamed reasoning during direct-model tool use, overflowing wide chat tables, and unreliable persistence for some settings and scoped tools."
          ]
        }
      }
    },
    {
      "version": "0.8.19",
      "releasedAt": "2026-08-11",
      "notes": {
        "zh-CN": {
          "features": [
            "新增简体中文与英文界面，可在设置中即时切换并跟随系统语言。",
            "新增统一的文档解析中心，为聊天附件和知识库导入提供原生文本提取、PDF 页面处理与真实文件诊断。",
            "新增本地 PP-OCRv6 Tiny、Small 和 Medium 模型，支持校验下载、离线识别以及受管 ZIP 导入和导出。",
            "扩展离线语音模型，新增中英与中粤英 Paraformer，以及 Whisper Small 和 Medium 多语言档位。",
            "增强魔法笔记，支持富文本、图片、视频、附件、待办状态和可配置的 AI 评论方式。",
            "支持为每个项目设置新对话的默认 Runtime。",
            "扩展直连模型工具与文档处理能力，增加联网搜索、网页读取、附件解析进度和当前系统时间上下文。",
            "新增首次启动版本更新说明，按当前界面语言展示且每个版本仅自动显示一次。"
          ],
          "fixes": [
            "修复 Execute 模式下内置 OpenCode 和 Continue 仍可能阻止已授权工具的问题。",
            "修复工具失败信息重复显示、已恢复的 OpenCode 响应仍被判定失败，并仅为最近一次失败保留重新编辑入口。",
            "修复共享开关在部分设置布局中尺寸被文本输入样式覆盖的问题。"
          ]
        },
        "en-US": {
          "features": [
            "Added Simplified Chinese and English interfaces with instant switching in Settings and system-language support.",
            "Added a unified document parsing center for chat attachments and knowledge imports, with native text extraction, PDF page handling, and real-file diagnostics.",
            "Added local PP-OCRv6 Tiny, Small, and Medium models with verified downloads, offline recognition, and managed ZIP import and export.",
            "Expanded offline speech models with bilingual and Mandarin-Cantonese-English Paraformer options, plus Whisper Small and Medium multilingual tiers.",
            "Enhanced Magic Notes with rich text, images, videos, attachments, editable todo states, and configurable AI comment modes.",
            "Added a per-project default Runtime for new conversations.",
            "Expanded direct-model tools and document handling with web search, webpage reading, attachment parsing progress, and current system-time context.",
            "Added first-open release notes that follow the current interface language and appear automatically only once per version."
          ],
          "fixes": [
            "Fixed authorized tools still being blocked for bundled OpenCode and Continue in Execute mode.",
            "Fixed duplicate tool-failure messages, preserved recovered OpenCode responses, and limited the edit-and-retry action to the latest failed response.",
            "Fixed shared switches inheriting text-input dimensions in some settings layouts."
          ]
        }
      }
    }
  ]
}