25 Commits

Author	SHA1	Message	Date
iven	0bb526509d	fix(identity): 名字检测从 memory extraction 解耦 — 502 不再阻断面板刷新 ... agent/user name 检测之前放在 extractFromConversation().then() 回调里， memory extractor 502 时整个 .then() 跳过，名字永远不会更新。 现在名字检测独立执行，memory extraction 失败不影响面板刷新。	2026-04-23 14:01:05 +08:00
iven	394cb66311	fix(identity): 重构 agent 命名检测正则 — 覆盖"名称改为小芳"等表达 ... detectAgentNameSuggestion 从固定正则改为 trigger+extract 两步法， 10 个 trigger 模式覆盖中文/英文常见命名表达，stopWords 过滤误匹配。 同时修复 streamStore content 类型处理和 RightPanel 重复事件监听。	2026-04-23 13:13:40 +08:00
iven	b56d1a4c34	feat(chat): LLM 动态对话建议 — 替换硬编码关键词匹配 ... AI 回复结束后，将最近对话发给 LLM 生成 3 个上下文相关的后续问题， 替换原有的"继续深入分析"等泛泛默认建议。 变更: - llm-service.ts: 添加 suggestions 提示模板 + llmSuggest() 辅助函数 - streamStore.ts: SSE 流式请求 via SaaS relay，response.text() 一次性 读取避免 Tauri WebView2 ReadableStream 兼容问题，失败降级到关键词 - chatStore.ts: suggestionsLoading 状态镜像 - SuggestionChips.tsx: loading 骨架动画 - ChatArea.tsx: 传递 loading prop	2026-04-23 11:41:50 +08:00
iven	08812e541c	fix(identity): 接通身份信号提取与持久化 — 对话中起名跨会话记忆 ... 根因: 记忆提取管道(COMBINED_EXTRACTION_PROMPT)提取5种画像信号 但无身份信号(agent_name/user_name)，不存在从对话到AgentConfig.name 或IdentityFiles的写回路径。 修复内容: - ProfileSignals 增加 agent_name/user_name 字段 - COMBINED_EXTRACTION_PROMPT 增加身份提取指令 - parse_profile_signals 解析新字段 + 回退推断 - GrowthIntegration 存储身份信号到 VikingStorage - post_conversation_hook 写回 soul.md + emit Tauri 事件 - streamStore 规则化检测 agent 名字并更新 AgentConfig.name - cold-start-mapper 新增 detectAgentNameSuggestion 链路: 对话→提取→VikingStorage→hook写回soul.md→事件→前端刷新	2026-04-23 09:20:35 +08:00
iven	58ff0bdde7	fix(kernel,desktop): Core Chain Hardening 穷尽审计 7 项修复 ... 审计发现 1 CRITICAL + 4 HIGH + 4 MEDIUM + 4 LOW, 修复如下: CRITICAL: - TS seam 测试改为 JSON round-trip 验证 (12 测试覆盖 10 事件类型) HIGH: - post_conversation_hook 拦截路径 driver=None 加 debug 日志 - schedule intercept channel send 失败回退 LLM (return Ok(None)) MEDIUM: - DeltaBuffer.flush() 先 mutation 再 clear, 防止异常丢数据 - ModelsAPI.tsx 去重: 改用 model-config.ts 导出 (消除 2 函数+1 接口+2 常量) - boot_with_driver docstring 记录跳过 agent 恢复 TypeScript 0 错误, Rust 76 kernel 测试通过, TS 12 seam 测试通过	2026-04-21 23:30:08 +08:00
iven	191cc3097c	refactor(desktop): streamStore sendMessage 拆分 Phase 3 — 提取 DeltaBuffer+4 Handler ... - DeltaBuffer 类: ~60fps 文本/思考增量缓冲，替换内联 textBuffer/thinkBuffer - createToolHandler: 工具步骤管理 (toolStart/toolEnd + artifact 自动创建) - createHandHandler: Hand 能力消息生成 - createSubtaskHandler: 子任务状态映射 - createCompleteHandler: 完成回调 (token 统计+记忆提取+反思+建议) - sendMessage 内联回调从 ~350 行缩减到 ~130 行 (-63%) - TypeScript 类型检查通过, 8 个 seam 测试通过	2026-04-21 23:03:04 +08:00
iven	b2908791f6	fix(desktop): Tauri 端找碴验证 7 项修复 — 消息泄漏/UUID暴露/错误友好化 ... B15/B11: streamStore onAgentStream 添加 activeRunId 过滤，移除降级匹配， hand/workflow 消息追加前验证 runId 归属；chatStore 切换/新建对话时 先 cancelStream 终止旧流；ChatArea hand-execution-complete 事件 添加 isStreaming 守卫 B4/B5: ChatArea 模型列表过滤 embedding 模型，provider 设为 undefined 隐藏 UUID B2/B3: streamStore onError 添加 formatUserError 函数，将原始 JSON 错误转换为中文友好提示 B1: SuggestionChips onSelect 延迟调用 handleSend 自动发送建议 fix(runtime): test_util.rs with_error 添加 mut self，with_stream_chunks 移除多余 mut fix(saas): lib.rs 添加 Result/SaasError re-export	2026-04-21 20:29:47 +08:00
iven	f2917366a8	fix(growth,kernel,runtime,desktop): 50 轮功能链路审计 7 项断链修复 ... P0 修复: - B-MEM-2: 跨会话记忆丢失 — 添加 IdentityRecall 查询意图检测， 身份类查询绕过 FTS5/LIKE 文本搜索，直接按 scope 检索全部偏好+知识记忆； 缓存 GrowthIntegration 到 Kernel 避免每次请求重建空 scorer - B-HAND-1: Hands 未触发 — 创建 HandTool wrapper 实现 Tool trait， 在 create_tool_registry() 中注册所有已启用 Hands 为 LLM 可调用工具 P1 修复: - B-SCHED-4: 一次性定时未拦截 — 添加 RE_ONE_SHOT_TODAY 正则匹配 "下午3点半提醒我..."类无日期前缀的同日触发模式 - B-CHAT-2: 工具调用循环 — ToolErrorMiddleware 添加连续失败计数器， 3 次连续失败后自动 AbortLoop 防止无限重试 - B-CHAT-5: Stream 竞态 — cancelStream 后添加 500ms cancelCooldown， 防止后端 active-stream 检查竞态	2026-04-20 09:43:38 +08:00
iven	24b866fc28	fix(growth,runtime,desktop): E2E 验证 4 项 Bug 修复 ... P1 BUG-1: SemanticScorer CJK 分词缺失导致 TF-IDF 相似度为 0 - 新增 CJK bigram 分词: "北京工作" → ["北京","京工","工作","北京工作"] - 非CJK文本保持原有分割逻辑 - 3 个新测试: bigram 生成 + 混合文本 + CJK 相似度>0 P1 BUG-2: streamStore lifecycle:end 未记录 token 使用量 - AgentStreamDelta 增加 input_tokens/output_tokens 字段 - lifecycle:end 处理中检查并调用 addTokenUsage P2 BUG-3: NlScheduleParser "X点半" 解析为整点 - 所有时间正则增加可选的 (半) 捕获组 - extract_minute 辅助函数: 半 → 30 P2 BUG-4: NlScheduleParser "工作日每天" 未转为 1-5 - RE_WORKDAY_EXACT 支持 (每天|每日)? 中缀 - try_workday 优先级提升至 try_every_day 之前 E2E 报告: docs/E2E_TEST_REPORT_2026_04_19.md 测试: 806 passed / 0 failed (含 9 个新增测试)	2026-04-20 00:07:07 +08:00
iven	cb9e48f11d	refactor(hands): 移除空壳 Hand — Whiteboard/Slideshow/Speech (Phase 5) ... 删除 3 个仅含 UI 占位的 Hand，清理 Rust 实现与前端引用： - Rust: whiteboard.rs(422行) + slideshow.rs(797行) + speech.rs(442行) - 前端: WhiteboardCanvas + SlideshowRenderer + speech-synth + 相关类型/常量 - 配置: 3 个 HAND.toml - 净减 ~5400 行，Hands 9→6(启用) + Quiz/Browser/Researcher/Collector/Clip/Twitter/Reminder	2026-04-17 19:55:59 +08:00
iven	1309101a94	fix(ui): Agent 面板信息不随对话更新 — 事件时序 + clones 刷新 ... - streamStore: zclaw:agent-profile-updated 事件从记忆提取前改为 .then() 后触发 - RightPanel: profile 更新事件中新增 loadClones() 刷新 selectedClone 数据	2026-04-16 22:57:32 +08:00
iven	02a4ba5e75	fix(desktop): 替换 require() 为 ES import — 修复生产构建崩溃 ... - connectionStore: 2 处 require() → loadConversationStore() 异步预加载 + 闭包引用 - saasStore: 1 处 require() → await import()（logout 是 async） - llm-service: 1 处 require() → 顶层 import（无循环依赖） - streamStore: 移除重复动态导入，统一使用顶层 useConnectionStore - tsc --noEmit 0 errors	2026-04-15 00:47:29 +08:00
iven	e0eb7173c5	fix: 三端联调 P1 修复 — API密钥页崩溃 + 桌面端401恢复 + 用量统计全零 ... P1-03: vite.config.ts proxy '/api' → '/api/' 加尾部斜杠， 防止前缀匹配 /api-keys 导致 SPA 路由崩溃 P1-01: kernel_init 增加 api_key 变更检测（token 刷新后自动重连）， streamStore 增加 401 自动恢复（refresh token → kernel reconnect）， KernelClient 新增 getConfig() 方法 P1-02: /api/v1/usage 总计改从 billing_usage_quotas 读取 （authoritative source，SSE 和 JSON 均写入）， by_model/by_day 仍从 usage_records 读取	2026-04-14 22:02:02 +08:00
iven	4c3136890b	fix: 三端联调测试 2 P0 + 6 P1 + 2 P2 修复 ... P0-1: SaaS relay 模型别名解析 — "glm-4-flash" → "glm-4-flash-250414" (resolve_model) P0-2: config.rs interpolate_env_vars UTF-8 修复 (chars 迭代器替代 bytes as char) + DB 启动编码检查 + docker-compose UTF-8 编码参数 P1-3: UI 模型选择器覆盖 Agent 默认模型 (model_override 全链路: TS→Tauri→Rust kernel) P1-6: 知识搜索管道修复 — seed_knowledge 创建 chunks + 默认分类 (seed/uploaded/distillation) P1-7: 用量限额从当前 Plan 读取 (非 stale usage 表) P1-8: relay 双维度配额检查 (relay_requests + input_tokens) P2-9: SSE 路径 token 计数修复 — 流结束检测替代固定 500ms sleep + billing increment	2026-04-14 00:17:08 +08:00
iven	b3f7328778	feat(ui): '我眼中的你' 双源渲染 — 静态Clone + 动态UserProfile ... - RightPanel 增加 userProfile state + fetch 逻辑 - 对话结束后通过 CustomEvent 触发画像刷新 - UserProfile 字段: 行业/角色/沟通偏好/近期话题 桥接 identity 系统 → 前端面板。	2026-04-11 12:51:28 +08:00
iven	9442471c98	fix(relay): send conversation history to SaaS relay (BUG-008) ... SaaS Relay was sending only the current message without conversation history, giving LLM no context from previous turns. Root cause: streamStore passed only `content` string to chatStream(), and saas-relay-client hard-coded a single-element messages array. Fix: - GatewayClient.chatStream() opts: add `history` field - streamStore: extract last 20 messages as history before calling chatStream - saas-relay-client: build messages array from history + current message	2026-04-09 22:41:56 +08:00
iven	e6eb97dcaa	perf(relay): full-chain optimization — key pool, model sync, SSE stream ... Phase 1 (Key Pool correctness): - RPM: fixed-minute window → sliding 60s aggregation (prevents 2x burst) - Remove fallback-to-provider-key bypass when all keys rate-limited - SSE semaphore: 16→64 permits, cleanup delay 60s→5s - Default 429 cooldown: 5min→60s (better for Coding Plan quotas) - Expire old key_usage_window rows on record Phase 2 (Frontend model sync): - currentModel empty-string fallback to glm-4-flash-250414 in relay client - Merge duplicate listModels() calls in connectionStore SaaS path - Show ModelSelector in Tauri mode when models available - Clear currentModel on SaaS logout Phase 3 (Relay performance): - Key Pool: DashMap in-memory cache (TTL 5s) for select_best_key - Cache invalidation on 429 marking Phase 4 (SSE stream): - AbortController integration for user-cancelled streams - SSE parsing: split by event boundaries (\n\n) instead of per-line - streamStore cancelStream adapts to 0-arg and 1-arg cancel fns	2026-04-09 19:34:02 +08:00
iven	5c6964f52a	fix(desktop): error response improvements — content, retry, model selector ... P1: onError callback now sets content to error message instead of empty string. Previously API errors (404/429) produced empty assistant messages with only a visual error badge — now the error text is persisted in message content. P3: Retry button now re-sends the preceding user message via sendToGateway instead of copying to input. Works for both virtualized and non-virtualized message lists. Removed unused setInput prop from MessageBubble. Also hides model selector in Tauri runtime (SaaS token pool routes models).	2026-04-09 18:52:27 +08:00
iven	0b0ab00b9c	fix(chat): prevent React Maximum update depth exceeded during streaming ... Root cause: Each LLM delta (text/thinking) triggered a synchronous setState via updateMessages → chatStore.setState. With Kimi thinking model emitting many deltas per frame, this caused a React render storm that hit the maximum update depth limit. Fix (two-layer approach): 1. streamStore: Buffer text/thinking deltas locally and flush to store via setTimeout(0), batching multiple deltas per frame 2. chatStore: Microtask batching in injectChatStore.updateMessages to coalesce rapid successive updates Verified: 2-round conversation (4 messages) with Kimi thinking model completes without crash. Previously crashed 100% on 2nd message.	2026-04-08 14:47:43 +08:00
iven	02c69bb3cf	fix: subagent unique ID matching + AgentState serialization + pre-existing TS errors ... - S-3: Thread task_id (UUID) through all 6 layers (LoopEvent → StreamChatEvent → kernel-types → gateway-client → streamStore) so subtasks are matched by ID, not description string - AgentState: Add #[serde(rename_all = "lowercase")] to fix PascalCase serialization ("Running" → "running"), update frontend matcher - S-1: Remove unused onClose prop from ArtifactPanel + ChatArea call site - Fix hooks/index.ts: remove orphaned useAutomationEvents re-exports (module deleted) - Fix types/index.ts: remove orphaned automation type/value re-exports (module deleted) - Fix ChatArea.tsx: framer-motion 12 + React 19 type compat — use createElement + explicit any return type to avoid unknown-in-JSX-child error Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-06 22:30:16 +08:00
iven	9871c254be	feat: sub-agent streaming progress — TaskTool emits real-time status events ... - Rust: LoopEvent::SubtaskStatus variant added to loop_runner.rs - Rust: ToolContext.event_sender field for streaming tool progress - Rust: TaskTool emits started/running/completed/failed via event_sender - Rust: StreamChatEvent::SubtaskStatus mapped in Tauri chat command - TS: StreamEventSubtaskStatus type + onSubtaskStatus callback added - TS: kernel-chat.ts handles subtaskStatus event from Tauri - TS: streamStore.ts wires callback, maps backend→frontend status, updates assistant message subtasks array in real-time	2026-04-06 13:05:37 +08:00
iven	15a1849255	feat: auto-create artifacts from file_write tool output ... When the agent writes files via the file_write tool, artifacts are now automatically created in the artifact panel. The file extension determines the artifact type (code/markdown/text) and syntax highlighting language. This connects the existing ArtifactPanel UI to actual tool output data.	2026-04-06 12:50:48 +08:00
iven	cb140b5151	feat: DeerFlow 2.0 core capabilities — Phase 1.0 + 1.1 ... Phase 1.0 — Butler Mode UI: - Hide "自动化" and "技能市场" entries from sidebar navigation - Remove AutomationPanel and SkillMarket view rendering from App.tsx - Simplify MainViewType to only 'chat' - Main interface is now: chat + conversation list + detail panel only Phase 1.1 — Mode Differentiation: - Add subagent_enabled field to ChatModeConfig (Rust), StreamChatRequest (Tauri), gateway-client, kernel-client, saas-relay-client, and streamStore - TaskTool is now only registered when subagent_enabled=true (Ultra mode) - System prompt includes sub-agent delegation instructions only in Ultra mode - Frontend transmits subagent_enabled from ChatMode config through the full stack This connects the 4-tier mode selector (Flash/Thinking/Pro/Ultra) to actual backend behavioral differences — Ultra mode now truly enables sub-agent delegation.	2026-04-06 12:46:43 +08:00
iven	769bfdf5d6	fix(desktop): 修复 10 个 agent 对话测试发现的缺陷 ... - BUG-001+002 P0: 模型选择器合并 SaaS 可用模型列表 - BUG-003 P1: 修复 relay 错误消息重复显示 - BUG-005 P1: 设置页面显示实际连接模式和地址 - BUG-006 P2: 统一 UI 语言为中文 - BUG-009 P3: 错误时隐藏建议按钮 - BUG-010 P3: 密码切换按钮添加 aria-label Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-04 01:31:28 +08:00
iven	0a04b260a4	refactor(desktop): ChatStore structured split + IDB persistence + stream cancel ... Split monolithic chatStore.ts (908 lines) into 4 focused stores: - chatStore.ts: facade layer, owns messages[], backward-compatible selectors - conversationStore.ts: conversation CRUD, agent switching, IndexedDB persistence - streamStore.ts: streaming orchestration, chat mode, suggestions - messageStore.ts: token tracking Key fixes from 3-round deep audit: - C1: Fix Rust serde camelCase vs TS snake_case mismatch (toolStart/toolEnd/iterationStart) - C2: Fix IDB async rehydration race with persist.hasHydrated() subscribe - C3: Add sessionKey to partialize to survive page refresh - H3: Fix IDB migration retry on failure (don't set migrated=true in catch) - M3: Fix ToolCallStep deduplication (toolStart creates, toolEnd updates) - M-NEW-2: Clear sessionKey on cancelStream Also adds: - Rust backend stream cancellation via AtomicBool + cancel_stream command - IndexedDB storage adapter with one-time localStorage migration - HMR cleanup for cross-store subscriptions	2026-04-03 00:24:16 +08:00