Add a supervisor text-model selector using existing connections and a per-conversation story graph switch above knowledge settings. Hide the switch when Supervisor is disabled and enforce conversation preferences at local and remote tool execution boundaries.
Release note: Choose a dedicated supervisor model and control whether each conversation can query existing stories and experiences. Enabling graph access does not start a review or automatically insert context.
Main module evaluation drops from ~420 ms to ~330 ms. Types stay as import type; the DeepSeek Harness chunks stay static because the protocol chunk shares code with Main startup paths.
Runtime creation, MCP launches, Obsidian, native clients and the tool-environment IPC await the initialization promise instead of startup awaiting it before the window path. A failure still fails startup with the same diagnostic.
Configuration apply previously required a separate native confirmation even in Agent Runtime Execute mode, and failed in clients without an approval callback. Use existing work-mode and tool authorization instead, preserving Ask read-only access and request-scoped plan validation. Update regression tests, tool descriptions and documentation.
Validation: 267 focused tests passed; typecheck and changed-file lint passed. Full tests timed out with failures in other modules; repository lint reported errors in unchanged files.
Release note: Execute 模式下通过 Agent Runtime 修改 GoodBuddy 配置不再额外弹出确认框，Ask 模式仍保持只读。
Expose save_image across direct model, MCP gateway, DeepSeek Harness and remote Agents (written on the remote host via chunked reads), and show the image MCP tool list in settings.
- Segment streaming Markdown so completed blocks are reused and only the growing tail is re-parsed; fall back to whole-document parsing for document-wide constructs
- Memoize conversation sidebar rows and cache per-conversation derived activity, ordering and pending approvals
- Activity page loads on entry, refreshes every 15 s while visible, and has a single header refresh button
- Include pending settings, model download source, device sharing and knowledge gateway changes in the working tree
- Add Electron streaming input benchmark and regression tests
Add an opt-in persistent frosted glass setting for the top bar and runtime checklist, with chat scrolling behind the header while preserving scroll anchors and window controls.
Retry transient local MCP initialization failures once after cleanup, include redacted connection diagnostics in feedback, and index lifecycle events to reduce execution statistics scans.
Release note: Add optional frosted glass appearance with chat visible behind the title bar; improve local tool connection recovery and diagnostic feedback.
Keep the notes list beside the open note, compact the detail view with a full-width sticky toolbar, and make Back to overview return to a full-width all-notes grid. Add manual note tags with AND filtering, a tag manager for rename, merge and delete, schema 51 tag tables, and tag support in the note Agent tools.
Use gbd-/gbc- prefixes with a 6-char conversation hash instead of 20-char per-request hashes, lease per-conversation slots so parallel requests never share a name, and expose gateway custom tools as <server hash>_<name> within the 64-char provider limit.
Every 'Open in terminal' launch for OpenCode copied the bundled plugin dependency tree (about 3,600 files, 52 MB) into a temporary directory, adding 2.2-2.4 s on Windows before the TUI started. The tree is now copied once into terminal/opencode-config under user data and reused until its bundled version marker changes. Skills, Ask-mode plugins and model configuration are still generated per launch and cleaned up on close. OpenCode's own TUI startup (about 3 s) is unchanged; the remote Agent path is unaffected.
Release note: 在终端中打开 OpenCode 时不再每次复制插件依赖，第二次及以后的打开速度明显加快。
- Build runtime history per turn: a cancelled or failed reply keeps its partial
text plus an interruption marker (or a marker-only placeholder), so its user
message is never left unanswered and merged into the next request.
- Use the same history for send, manual compaction and Main compact validation.
- OpenCode: keep a cancelled prompt in the reused native session and prefix the
next turn with a one-time synthetic interruption notice; failed prompts are
still removed.
- Add an opt-in DeepSeek live test covering direct model and OpenCode.
- Store runtime selection as { provider?, model? } layers and resolve them per
field (conversation -> project -> global); drop the legacy auto provider.
- Migrate stored selections (database v49) and clear deleted profile ids.
- Re-resolve selections in Main per request and at queue dispatch.
- Add ProjectRuntimeSelector and a two-column RuntimeModelPicker; manage
link opens model connections; restore per-runtime default model tabs.
- Stop duplicating trusted instructions and history across OpenCode,
Continue and DeepSeek Harness runtimes.
- Document the layered design, global provider removal plan and planned
purpose-specific models.
Add a System tasks tab to run history that splits background usage (supervisor, auto-supervision, magic notes, knowledge base, delegation) from conversation usage, with a manual refresh button and automatic refresh when usage is recorded. Record previously dropped usage from supervisor reviews, image regeneration, knowledge graph extraction, and remote embeddings.
Local direct-model text conversations could not use manual context compaction,
and disabling automatic compression prevented saved summaries from being reused.
Restore the composer action regardless of the automatic setting or threshold,
use the selected summary connection, and persist summaries for subsequent
requests while retaining recent turns and full conversation history.
Image connections, remote channel conversations, and SSH projects remain outside
the local direct-model compaction workflow. Add focused regressions, a local
Electron end-to-end test, and an opt-in two-request live probe.
Validation: staged snapshot passed typecheck, changed-file ESLint, 13 focused
tests, and the Electron end-to-end test with two local HTTP fixture requests.
No additional real-model calls were made during commit verification.
Release note: 修复本地直连文本模型无法手动压缩上下文的问题。现在关闭自动压缩或尚未达到触发阈值时，也可点击“压缩上下文”；摘要会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。
Remove the fixed gateway capability lifetime so active requests retain knowledge, notes, browser, configuration, and custom MCP access. Preserve token validation and request cancellation or cleanup revocation, and replace the misleading knowledge expiry error with a tool authorization error.
Remove native-client clock overrides and cover long-running scoped and HTTP MCP calls. Validation: 34 focused tests passed; changed-file lint passed. Repository typecheck and lint remain blocked by unrelated manual-compaction and sidebar probe files.
Release note: Long-running Agent requests no longer lose tool access after ten minutes, including when the knowledge-base switch is off. Request cancellation and completion still revoke access.
Expose saved decisions, timelines and source evidence through read-only Story Graph tools across Model, Harness and local/remote OpenCode and Continue. Supervisor enablement and runtime assignments govern discovery and calls; reads do not invoke models or modify stored evidence. Remote use requires matching Desktop and Agent changes.
Make Continue discussion show the complete editable message with a separated destination and shared action layout. Sending uses the visible draft without reattaching removed context.
Validation: existing focused MCP, runtime, controlled Linux Host, discussion component and Electron layout checks plus typechecks; staged diff check passed. Cross-scope fact reuse, historical reconstruction and full package acceptance remain deferred.
Release note: Agents can consult saved supervisor decisions and their sources when continuing work. Continue discussion now lets you review and edit the complete message before sending.
Run conversation capture acceptance in a visible focused Electron window
and report concrete focus state on failure. Trigger Continue pre-spawn
cancellation at its existing environment callback instead of racing a timer.
Product implementation and approved bilingual note text remain unchanged.
Include the user's companion article updates and three reviewed draft images
as authorized. Keep their draft status and unverified build provenance clear.
Set the release date to September 28 and record the final real Linux result:
5345 passed, 51 skipped; Windows focused, typecheck, lint, and notes passed.
Continue still displayed Checking for updates even when version lookup was disabled. Skip the update service before it emits checking state and retain the lower-level lookup guard in the shared local and Agent adapter.
Add a regression for disabled and normal update paths and record real terminal and production Electron validation. Seven real model requests returned HTTP 200; Linux validation covered startup only. The reported UI freeze was not reproduced, full Agent/model acceptance remains open, and full-suite failures are documented.
Release note: 修复 Continue 禁用自动更新后仍显示 Checking for updates 的问题，终端现在保持正常空闲状态。
DS Web and remote Agent launch paths omitted shared telemetry opt-out settings. Reuse one environment definition for local launches, remote runtime profiles, and the remote model bridge helper while preserving existing update controls and user-requested networking.
Validation: 16 focused tests passed at commit time. Prior local real-model validation passed; modified Agent real-Host validation and background network monitoring remain outstanding.
Release note: 补齐 DS Web 与远程 Runtime 启动时的遥测关闭配置，保留现有自动更新关闭设置，不影响模型、MCP 和用户主动联网请求。
Add composer shortcuts for local Continue and OpenCode terminals and the official DeepSeek Harness Web client, using the current project and model bridge. Manage browser service reuse, stop actions and terminal cleanup. Upgrade DeepSeek Harness to 0.1.7-rc.2 and Cordis to 4.0.4.
Include the coordinated IPC handler extraction and related tests and documentation. Remote native-client coordinator wiring, managed Node consolidation, Ask MCP mappings and full desktop/package acceptance remain unfinished; this commit records the implemented local workflow and supporting SSH groundwork.
Validation: typecheck and lint passed; 267 focused tests passed and 5 skipped. No new real-provider calls were made for this commit. Prior real-model evidence and remaining acceptance work are documented in the feature progress files.
Release note: 本地项目选择 Continue 或 OpenCode 后，可从输入框下方打开原生终端；选择 DS 后，可打开使用当前项目和模型的官方 Web 客户端。客户端使用独立会话，DS 网页关闭后服务继续运行，可在 GoodBuddy 中停止；远程入口尚未完成。
Let users retain useful assistant replies or complete conversations without
leaving their chat workflow. Preview and edit captured text, choose an existing
note or atomically create a note with its first entry, and return to the source
conversation or message from saved records.
Add an application-scoped Notes workbar tab with title/body search and quick
text appends. Preserve drafts across conversation switches and guard closing,
replacement and disabling. Schema 48 stores optional source metadata without
rewriting file-backed bodies; source deletion retains the saved note. Older
clients cannot reopen an upgraded database without a pre-upgrade backup.
Release note: 可将助手回答或完整会话预览、编辑后加入魔法笔记，并从笔记返回原会话或消息。右侧快速笔记支持搜索和追加文字，切换会话保留草稿；采集本身不调用模型，保存后自动评论仍遵循已有设置。
Validation: isolated HEAD-based candidate passed 22 files / 930 focused tests,
plus 20 gateway tests after including its create-result fixture adjustment.
Real Electron App/preload/IPC/SQLite acceptance passed for all 502 historical
messages, exact-entry/source navigation, native keyboard input in a 214px
panel, reopen and source-deletion retention, with zero capture model calls.
Repository lint and web TypeScript passed. Cached whitespace and secret-pattern
checks passed; native-client changes and unrelated IPC extraction are excluded.
Validation limits: the isolated source tree reuses installed node_modules,
where a parallel task upgraded DSH from the committed 0.1.2-rc.1 to 0.1.7-rc.2.
Full typecheck reports 14 DSH errors in three unchanged files; the full-suite
attempt with --bail=1 also fails in deepseek-harness-host.test.ts. These are not
all-green repository or release-package results. Earlier execution-time full
suite failures and fixture repairs remain recorded in the feature progress doc.
Users can now read, search, and update existing local Obsidian notes through
built-in tools without configuring an external MCP server or Obsidian plugin.
Add vault discovery, an optional folder scope, and independent test, save,
enablement, and runtime assignment controls in MCP settings.
Bundle MCPVault 0.16.0 and its runtime dependencies, exposing 19 shared tools
to direct models and managed local OpenCode and Continue conversations.
Obsidian is disabled by default. Ask remains read-only; Execute permits
writes. Scope and assignment changes revoke existing Obsidian grants.
DeepSeek Harness and remote execution remain unsupported.
Validate the isolated staged snapshot with its own dependency installation:
typecheck, lint, production build, and 540 tests across 14 files passed.
No paid model calls were repeated. Physical-offline, installer, and other
platform acceptance remain pending as documented in the feature record.
Release note: Connect local Obsidian vaults in MCP settings to find notes and
write conversation results back to them. Test all registered vaults or a
chosen folder before enabling access; no Obsidian plugin is required.
Replace the restricted workspace_rg parameter mapping with native args and optional cwd. Preserve rg output and exit codes, and reuse process cancellation and output_read pagination so large search results remain accessible. Ask remains read-only within the workspace; Execute uses current-account permissions.
Validation: focused regression 169 passed, 1 skipped; typecheck and lint passed. Real DeepSeek Ask and Execute pagination scenarios passed with six counted requests in the successful run; an earlier failed fixture run has no retained request count. Full-suite validation is not green: the recorded run had unrelated failures and the latest run timed out without a summary. The remote Agent does not use this local execution path.
Release note: 直连模型可使用 ripgrep 原生参数搜索文件并保留原生结果，大输出可通过 output_read 完整续读。Ask 保持工作区只读，Execute 使用当前账号权限。
Wait for model delivery acknowledgements before completing remote prompts and preserve completed outcomes when idle Runtime cleanup fails. Explicit messages after an unknown outcome now open a fresh binding with history instead of reusing the failed session; recovery remains attach-only.
Run obsolete payload cleanup through the current Agent after activation, retaining history and unresolved process ownership. Live legacy daemons and shared Runtime versions remain deferred. Includes focused regressions, local OpenCode simulation, and isolated Linux validation evidence. Typecheck and lint passed; the full test run timed out.
Release note: Remote conversations can continue after an uncertain result, and completed answers no longer fail solely because Runtime cleanup fails. Updates remove verified inactive Agent payloads and owned orphan processes while retaining active or uncertain installations and conversation history.
Linux release CI could not start OpenCode because the isolation test assumed a platform-specific npm executable name. Reuse the production runtime path resolver, which matches the installed bin/opencode.exe on all platforms.
Validation: nine focused tests, typecheck, lint, and bilingual release-note verification passed. Product behavior, versions, and release notes remain unchanged; main-branch CI must pass before tagging.
Address Desktop CI 35849362694 failures with isolated OpenCode config dependencies, cleared IPC lookup counts, Supervisor default-off expectations, and polling-safe UI fixtures and timeouts.
Record the failed candidate CI and follow-up validation: 14 focused tests, typecheck, and lint passed. Product code, versions, and bilingual release notes are unchanged. The revised candidate requires approval and successful main-branch CI before tagging.
Remote tasks could leave idle Runtime processes behind, and replaced Agent
installations lacked a drain-and-exit lifecycle. Keep the current Agent
resident, let displaced Agents finish existing work and reconnect for its
results, then retire them without restarting obsolete installations.
Reclaim Runtime processes when their last active task finishes, preserve
pending questions and shared active tasks, and restore conversation history
when the next turn starts. Remove the Remote Projects technical preview label
from settings, errors, tests, and current documentation; the feature remains
disabled by default.
Validation recorded during implementation: 213 lifecycle tests and 28 focused
UI/IPC tests passed, as did SettingsPanel and lint. Linux x64 real-Host checks
used 8 provider requests. Broader tests and final typecheck encountered
unrelated concurrent changes; details and coverage limits are documented.
Release note: 远程任务结束后自动释放空闲运行进程，下一轮对话恢复历史；更新后的旧 Agent 等待已有任务完成再退出，当前 Agent 保持常驻。远程项目移除“技术预览”标识，仍默认关闭。
Save only changed activity records instead of rewriting the entire history snapshot. Migrate existing history to schema 44 while preserving order and duplicate IDs.
Run unit tests with up to four workers and keep the configured integration phase serial. Include activity-history I/O, storage-upgrade, settings, and OpenCode isolation regression coverage.
Validation: 4 related test files / 123 tests passed; typecheck and lint passed.
Release note: Reduce disk writes while agent activity updates stream in, preserving existing activity history.
Keep paragraph formatting and checklist todos when removing canvas pages,
and preserve accepted undo/redo history after oversized input is rejected.
Merge restored queued text into the latest conversation draft and retain
image delivery when OCR returns image references and generated warnings only.
Deliver session MCP servers to bridged remote Continue Execute while keeping
Ask read-only and preserving local model-profile isolation. Add regression
coverage and update feature documentation. Correct stale conversation-limit
and forced-color assertions and the measured Electron layout test deadline.
Validation: 4,826 tests passed, 67 skipped; full typecheck and lint passed.
Current-source Linux x64 Execute/Ask/Execute MCP delivery passed, using 10
real text-provider requests and no image-provider requests. Image generation
itself was not exercised. Staged whitespace checks passed before commit.
Release note: 修复恢复队列时覆盖新输入、画布删页丢失清单格式及超限输入清空撤销历史的问题；图片 OCR 无正文时保留原图发送方式，远程 Continue Execute 正确保留会话 MCP 工具。
Magic Notes previously supported rich-text records without a paged drawing or
PDF annotation workflow. Add the integrated PeopleLib Fabric and Quill editor
with flowing text, pen/highlighter, object transforms, images, paper templates,
PDF backgrounds with native text extraction, and raster PDF export. Saves are
manual; undo/redo remains scoped to an annotation page or Quill mode.
Analyze drafts, saved canvases and canvas-source todos with page images and
text when the default model supports images, or an explicitly labelled
text-only fallback. Keep successful saves when automatic analysis fails and
prevent MCP plain-text replacement of canvas records.
Move entry bodies and binary assets into local note files with schema 38
migration and verified readback. SQLite retains metadata, indexes, revisions,
todos and comments. Backups must coordinate SQLite and the adjacent notes
directory while writes are paused or the app is closed; no one-click backup
action is included.
Show a continuous record stream with independent collapsible record index and
AI pane, real first-page thumbnails, and scroll-only record navigation. Open a
blank text/canvas composer by default and reset it after creation. Protect
unsaved drafts and establish normalized editor baselines after saving so clean
records do not prompt again when leaving.
Validation: latest recorded focused run passed 149 tests across five files;
typecheck, lint and git diff --check passed. The prior user-reported full run
passed 4,713 tests before later corrections and UI changes; it is not a full
validation of this final snapshot. Tests were not rerun for this commit, as
requested; independent systematic review follows the commit.
Release note: Create paged notes with handwriting, flowing text and PDF
annotations, browse records through thumbnails, and request AI comments with
an explicit text-only fallback. Existing notes migrate to local file storage;
backups now need both the database and notes directory.
Display native OpenCode and Continue checklists alongside scheduled tasks, preserving request ownership, explicit clears and saved progress. Bundle both remote runtimes and support managed Continue installation, model routing, cancellation and restart continuation.
Validation: real local and Linux remote model checks, full Electron remote Continue flows, typecheck and lint passed. Full suite: 4707 passed, 67 skipped, one OpenCode compaction failure passed targeted rerun.
Release note: Display native execution checklists above conversations and support managed remote Continue projects with saved progress and cancellation recovery.
Wait for OpenCode session cancellation before reusing a conversation, and replace the session after unconfirmed cancellation. Preserve terminal message state when stale streaming snapshots arrive so finished conversations return to Send.
Validated with full production Electron UI and the configured real text model: queued requests complete, cancellation followed by immediate resend completes, and Stop clears. Focused tests: 362 passed; typecheck and targeted lint passed. Live dispatch counting limitations are recorded in the feature evidence.
Release note: Consecutive OpenCode requests no longer inherit a previous cancellation, and finished conversations no longer retain the Stop button after stale updates.
Raise note_search limit to 100 while preserving the default and output budget. Return scoped-tool argument validation failures as tool errors so models can correct and retry instead of receiving MCP internal errors.
Validation: 64 focused tests, typecheck and changed-file lint passed. Full tests timed out with two heartbeat migration failures; full lint reports unrelated story-graph demo errors.
Release note: 笔记搜索单次最多支持 100 条结果；内置 MCP 参数错误会返回明确提示，便于模型纠正后重试。
Align remote ACP tool recovery with local runtime behavior when the model responds after a tool failure and completes normally. Preserve original errors and derive recovery state from committed remote events so reconnecting safely replays partial terminal events.
Validation: 461 focused tests, typecheck, lint, and shared Linux Host recovery scenarios passed. Full-suite validation has two pre-existing heartbeat migration fixture failures involving the pinned column.
Release note: 修复远端 Agent 在读取等工具失败后已正常回复，任务仍被误判为失败的问题，并修复断线重连时的工具状态恢复。
Links requesting a new window were silently blocked. Open them in managed browser tabs with the existing session restrictions, clean up script-closed tabs, and follow click-created tabs in subsequent direct and MCP browser tools.
Validation: 232 targeted tests and Windows Electron E2E passed; typecheck and lint passed. The full suite retains six failures outside the popup path.
Release note: Browser links that open a new page now open in a managed tab, and AI browser actions follow the new page after a click.
Add model-controlled image invocation and read-only runtime assignments. Route image tools through Main so accepted operations retain their originating conversation when chat runtimes change, with saved references, cancellation, and manual recovery.
Keep Node hashing outside browser-shared contracts to restore production builds. Validation includes the production build, typecheck, lint, focused regressions, real Host transport checks, and two successful real image requests. Full UI-driven runtime switching acceptance remains documented as pending.
Release note: 开启图片模型的会话调用后，聊天助手可调用图片生成与编辑工具，Runtime 分配自动管理。图片成果保存在原会话，可继续修改，并独立显示生成状态和取消入口。
Multiple projects could spawn duplicate heavyweight runtimes, while settings
changes and remote reconnection could cancel or strand native conversations.
Share compatible local OpenCode and DeepSeek Harness process owners and
Agent-owned ACP connections, retaining workspace routing, per-session modes,
model attribution, question ownership, targeted cancellation and idle cleanup.
Restore remote pending questions and terminal task projections across reconnect
and desktop restart, and preserve accepted ordinary requests during settings
replacement while cancelling expert-team parents with their replaced scheduler.
Extend regression coverage and Agent type checks. Record measured resource and
real-model validation results, including failures and limitations.
Also include the conversation media-generation design drafts (not an implemented
feature), documentation entry links and dedicated Linux test-directory guidance.
Desktop and Agent changes must be delivered together for shared process results.
Validation: typecheck and lint passed; staged content and documentation links
checked. Prior full regression and live-model evidence are recorded in progress
notes; no new full test suite or model calls were run for this commit.
Release note: 多项目会话现在复用兼容的 OpenCode 和 DeepSeek Harness 进程，减少重复启动并保持工作区、工具和取消隔离。修复远端问答在断线、重启及配置更新后的恢复与状态同步问题，Desktop 与 Agent 需配套更新。
Subscribe the open Magic Notes workspace to committed database changes from desktop and Agent/MCP writes. Preserve selections and unsaved drafts during refresh and external deletion.
File-level commit includes existing runtime edits in shared files; their remaining dependencies are still in the working tree.
Release note: Magic Notes now updates automatically after Agent/MCP edits while preserving your selection and unsaved drafts.
Show bounded operation summaries for Continue, DeepSeek Harness and direct-model tools, retaining input details across incremental updates. Use one canonical workspace path for DeepSeek Host startup and ACP sessions so equivalent Windows paths reuse a working Runtime.
Compare persisted subagent tool fields by value, release terminal task caches, and avoid rebuilding caches for duplicate remote events. Upgrade to schema 35 to recompact repeated schema-34 tool blocks while preserving events and provenance. Include regression coverage, validation records and temporary-resource test exclusions.
Validation: 4155 tests passed, 66 skipped; typecheck and lint passed. Full desktop real-model validation covered all three tool-summary paths and both Windows path spellings. Storage validation evidence includes local and remote tool workloads and real database upgrade checks.
Release note: Tool records now show paths, commands and search details across runtimes. Fix DeepSeek Windows workspace failures and repeated subagent history growth; schema 35 reclaims redundant history data while preserving execution records, and older clients cannot reopen the upgraded database.
Preserve OpenCode operation titles in main-session tool summaries, including permission waits, failures and interrupted calls. Keep title trimming and length limits aligned with subagent records.
Update tool display documentation and include the activity-menu wording in UI-DESIGN.md. Validation: 132 focused tests passed; main-process typecheck and changed-file lint passed. Full-suite checks were affected by concurrent activity-menu changes and a test timeout.
Release note: OpenCode main-session tool records now show file paths, commands or operation descriptions beside tool names, matching subagent records.
Native browser pages could cover application dialogs and overlapping menus
or notifications. Suspend only the viewport while covered and restore the
same active page afterward, including nested dialogs and terminal prompts.
Keep knowledge source controls at their full height, constrain long library
cards and titles, and update automatic names when switching remote targets
without overwriting edited or saved names. Restore packaged release-note
loading now that the history exceeds its previous size limit.
Add focused regressions and real Electron UI evidence. Clean up temporary
Harness fixture directories at test completion instead of accumulating them.
Release note: 修复内嵌浏览器遮挡弹窗和浮层、知识库来源选项被压缩及长名称显示异常；切换远端知识库时自动名称正确更新，并恢复更新说明显示。
Completed parallel conversations could leave buffered event connections open.
Chat and native context compaction now cancel their own subscriptions before
ending SDK iteration, without cancelling another conversation or its Server.
Strengthen cleanup-order regressions and verify real OpenCode parallel runs,
cancellation and compaction. The remote Agent uses ACP stdio and transcript
polling rather than this SDK event stream and needs no corresponding change.
Validation: 99 focused tests passed; the final candidate passed 4089 tests
with 66 skipped, typecheck and lint. No external model calls were made.
Cancelling a managed SSH prompt while answering a question could leave the conversation bound to its previous operation and reject the next message. Record the Agent cancellation outcome and reconcile and close the desktop binding before the next prompt.
Add two cancellation regressions and document real SSH/Electron question validation across eight theme and viewport combinations. Local queues and remote question routing are already included in 3b560c9.
Validation: prior focused checks passed 275 tests with 5 skipped; typecheck and lint passed. The final full run passed 4087 tests with 66 skipped and one packaging timeout; that packaging file separately passed 22 tests with 1 skipped. Staged diff whitespace check passed.
Release note: After cancelling a pending OpenCode question on a managed SSH task, you can send another message in the same conversation without an operation identity error.
Connect Dify, FastGPT and RAGFlow through read-only bindings, encrypted local credentials, shared retrieval and persistent citations. Configuration and removal affect GoodBuddy only.
Show active conversations across projects, retain older live work, and improve project dialog scrolling. Queue concurrent questions and route managed SSH OpenCode questions through the Agent. Preserve recoverable read/search errors and partial search results.
Document the Electron UI automation standard, real model and external knowledge validation, and remaining acceptance limits.
Validation: earlier full regression passed 4081 tests with 66 skipped; subsequent focused remote-question checks and real Linux evidence are recorded in feature progress. Precommit typecheck and lint passed; latest project dialog and shared UI suites passed 58 tests. The full suite was not rerun for the final dialog changes.
Release note: Use external knowledge in conversations, find active work across projects, and answer concurrent local or managed SSH OpenCode questions. Managed SSH questions require the updated Agent adapter.
Allow users to close leased browser tabs without cancelling the AI request or revoking its other tools. Closed-page operations report a recoverable error; explicit navigation creates a dedicated replacement tab for direct-model and MCP calls.
Add regression coverage and real Electron lifecycle checks, and update browser behavior documentation.
Release note: 浏览器标签页被 AI 请求占用时也可关闭，其他任务继续运行；再次导航会打开独立新页。
Image requests previously sent only the current prompt and attachments, so
follow-up edits could lose the previous image and conversation requirements.
Reuse the latest successful image from the existing artifact store together
with text history, including after reopening the app. Explicit attachments
take priority, and configured model IDs remain unchanged.
When image editing is explicitly unsupported, continue with text-based image
generation and retain a quiet message footer explaining the missing image
context. Keep genuine provider errors and cancellation visible. Add focused
regressions, persisted footer metadata, and aligned feature documentation.
Validated with four real sub-gpt-image-2 image requests and a simulated
unsupported-edit response. Image regressions, typecheck and lint passed;
the full working-tree run had two unrelated concurrent Runtime-menu failures.
Release note: 连续改图会自动带上上一张图片和文字上下文，重开对话后仍可继续修改。上游明确不支持编辑时仍可按文字生图，仅在回复底部轻提示未使用参考图片。
Restore Ask workspace reads and output paging without granting write access,
and let programming Subagents reuse their parent's browser page. Allocate
browser resources only when used, keep Agent pages aligned with workbar tabs,
recover released bindings, and remove unused automatic screenshot work.
Preserve inherited Runtime selections during remote usage persistence and
manual compression. Keep viable channel model fallbacks, display the actual
scheduled-task mode and Harness model source, and avoid rediscovery delays
when calling healthy MCP tools. Fix directory/history refresh, case-only
renames, rename-aware commit diffs, and FastGPT probe success reporting.
Add regression coverage and align feature documentation. The Agent-side
connection timeout fix requires deploying an updated Agent; existing Ask
read-only and Execute authorization boundaries remain unchanged.
Final validation: 3839 tests passed, 66 skipped; typecheck, lint, development
build, and diff checks passed. Isolated Electron UI/MCP workflows passed with
a loopback model fixture. Current-source Linux x64 Host checks passed with
three successful real model requests. Chromium loadURL rejection was not
reproduced in the UI; released-binding recovery has automated coverage.
Release note: 修复 Ask 读取项目文件、浏览器页面显示与恢复、文件管理及模型配置跟随中的不一致，并减少无用截图和 MCP 等待。定时任务显示实际工作模式，远程会话可正常保存继承 Runtime 的用量；远端连接超时修复需更新 Agent 后生效。
Account for Electron-as-Node when probing the current executable and update the Linux workspace registry scenario to validate the writable handle behavior shipped with remote file management.
mesalogo