diff --git a/frontend-web/public/runtime-config.js b/frontend-web/public/runtime-config.js
index 05e3fe6d..24cc025a 100644
--- a/frontend-web/public/runtime-config.js
+++ b/frontend-web/public/runtime-config.js
@@ -52,7 +52,7 @@ window.__CM_RUNTIME_CONFIG__ = {
 
   // 8BF 嵌入页专用任务系统接口。部署到其他系统时只需修改这一项，
   // 不影响原多智能体会商使用的任务详情接口。
-  VITE_8BF_TASK_API_BASE_URL: "https://ch1.b.sit.4.cn/zl_consumer",
+  VITE_8BF_TASK_API_BASE_URL: "https://ch1.b.uat.4.cn/zl_consumer",
 
   // 8BF 会商页右上角「返回」按钮的跳转地址（当前窗口 window.location.href）。
   // 部署后直接改 dist/runtime-config.js 即可，无需重新打包。留空则点击时提示未配置。
diff --git a/frontend-web/src/components/ai-elements/reasoning.tsx b/frontend-web/src/components/ai-elements/reasoning.tsx
index 2845edf6..d9c63e2a 100644
--- a/frontend-web/src/components/ai-elements/reasoning.tsx
+++ b/frontend-web/src/components/ai-elements/reasoning.tsx
@@ -165,7 +165,7 @@ export const ReasoningTrigger = memo(
 export type ReasoningContentProps = ComponentProps<
   typeof CollapsibleContent
 > & {
-  children: string;
+  children: ReactNode;
 };
 
 export const ReasoningContent = memo(
@@ -178,7 +178,11 @@ export const ReasoningContent = memo(
       )}
       {...props}
     >
-      <Streamdown {...reasoningPlugins}>{children}</Streamdown>
+      {typeof children === "string" ? (
+        <Streamdown {...reasoningPlugins}>{children}</Streamdown>
+      ) : (
+        children
+      )}
     </CollapsibleContent>
   ),
 );
diff --git a/frontend-web/src/components/workspace/agents/agent-detail-panel.tsx b/frontend-web/src/components/workspace/agents/agent-detail-panel.tsx
index a09653aa..c2c12ef1 100644
--- a/frontend-web/src/components/workspace/agents/agent-detail-panel.tsx
+++ b/frontend-web/src/components/workspace/agents/agent-detail-panel.tsx
@@ -2,6 +2,7 @@
 
 import {
   BotIcon,
+  BriefcaseBusinessIcon,
   CheckIcon,
   DatabaseIcon,
   GlobeIcon,
@@ -83,6 +84,7 @@ interface AgentDraft {
   /** 会商席位「技能识别强化」开关。 */
   seatSkillDirective: boolean;
   knowledgeBaseIds: string[];
+  jobResponsibilities: string;
 }
 
 function draftFromAgent(agent: Agent): AgentDraft {
@@ -97,6 +99,7 @@ function draftFromAgent(agent: Agent): AgentDraft {
     templateName: agent.template_name ?? "",
     seatSkillDirective: agent.seat_skill_directive ?? false,
     knowledgeBaseIds: agent.llmwiki_knowledge_base_ids ?? [],
+    jobResponsibilities: agent.job_responsibilities ?? "",
   };
 }
 
@@ -338,6 +341,8 @@ export function AgentDetailPanel({
       request.template_name = draft.templateName.trim();
     if (draft.seatSkillDirective !== (agent.seat_skill_directive ?? false))
       request.seat_skill_directive = draft.seatSkillDirective;
+    if (draft.jobResponsibilities.trim() !== (agent.job_responsibilities ?? ""))
+      request.job_responsibilities = draft.jobResponsibilities.trim();
     if (
       JSON.stringify(draft.knowledgeBaseIds) !==
       JSON.stringify(agent.llmwiki_knowledge_base_ids ?? [])
@@ -720,6 +725,39 @@ export function AgentDetailPanel({
       </section>
 
 
+      {/* ── 岗位职责 ───────────────────────────────────────────────────── */}
+      <Section
+        icon={<BriefcaseBusinessIcon className="size-3.5" />}
+        title="岗位职责"
+      >
+        {editing && draft ? (
+          <Textarea
+            value={draft.jobResponsibilities}
+            onChange={(e) =>
+              setDraft({ ...draft, jobResponsibilities: e.currentTarget.value })
+            }
+            placeholder="描述该智能体对应岗位的职责…"
+            className="min-h-20 text-sm"
+          />
+        ) : agent.job_responsibilities?.trim() ? (
+          <p className="text-muted-foreground text-sm leading-relaxed whitespace-pre-wrap">
+            {desensitize(agent.job_responsibilities)}
+          </p>
+        ) : canEdit ? (
+          <button
+            type="button"
+            onClick={beginEdit}
+            className="text-muted-foreground hover:text-foreground hover:border-border w-full rounded-lg border border-dashed px-3 py-4 text-center text-xs transition-colors"
+          >
+            <PlusIcon className="mr-1 inline size-3.5 align-[-2px]" />
+            添加岗位职责
+          </button>
+        ) : (
+          <p className="text-muted-foreground/60 text-sm">暂无岗位职责</p>
+        )}
+      </Section>
+
+
       {/* ── 常用问题 ───────────────────────────────────────────────────── */}
       {(showCommonQuestions || editing || questions.length > 0 || canEdit) && (
         <Section
diff --git a/frontend-web/src/components/workspace/messages/message-list-item.tsx b/frontend-web/src/components/workspace/messages/message-list-item.tsx
index b6945e1c..87d33005 100644
--- a/frontend-web/src/components/workspace/messages/message-list-item.tsx
+++ b/frontend-web/src/components/workspace/messages/message-list-item.tsx
@@ -6,10 +6,10 @@ import {
   ExternalLinkIcon,
   FileIcon,
   Loader2Icon,
-  PencilIcon,
-  SearchIcon,
-  SparklesIcon,
-} from 'lucide-react';
+  PencilIcon,
+  SearchIcon,
+  SparklesIcon,
+} from 'lucide-react';
 import {
   memo,
   useCallback,
@@ -17,10 +17,10 @@ import {
   useEffect,
   useMemo,
   useState,
-  type AnchorHTMLAttributes,
-  type ImgHTMLAttributes,
-  type ReactNode,
-} from 'react';
+  type AnchorHTMLAttributes,
+  type ImgHTMLAttributes,
+  type ReactNode,
+} from 'react';
 import rehypeKatex from 'rehype-katex';
 
 import { Loader } from '@/components/ai-elements/loader';
@@ -695,6 +695,17 @@ function MessageContent_({
     const withCitations = applyInlineCitations(rawContent ?? '', displaySources);
     return stripIncompleteCitationMarker(withCitations);
   }, [rawContent, isHuman, displaySources]);
+  const reasoningToDisplay = useMemo(() => {
+    if (!reasoningContent) return '';
+    const withCitations = applyInlineCitations(reasoningContent, displaySources, {
+      appendSummary: false,
+    });
+    return stripIncompleteCitationMarker(withCitations);
+  }, [reasoningContent, displaySources]);
+  const markdownRehypePlugins = useMemo(
+    () => [...rehypePlugins, [rehypeKatex, { output: 'html' as const }]],
+    [rehypePlugins],
+  );
   const parsedHumanContent = useMemo(
     () =>
       isHuman
@@ -724,23 +735,6 @@ function MessageContent_({
     );
   }
 
-  // Reasoning-only AI message (no main response content yet)
-  if (!isHuman && reasoningContent && !rawContent && !multiRoundTrace) {
-    return (
-      <AIElementMessageContent className={className}>
-        <Reasoning isStreaming={isLoading}>
-          <ReasoningTrigger />
-          <ReasoningContent>{reasoningContent}</ReasoningContent>
-        </Reasoning>
-        <MessageTokenUsage
-          enabled={tokenUsageEnabled}
-          isLoading={isLoading}
-          message={message}
-        />
-      </AIElementMessageContent>
-    );
-  }
-
   if (isHuman) {
     // Only the user's own question carries a delete affordance, and only once
     // it exists server-side (real, non-optimistic id). Deleting it cascades to
@@ -776,25 +770,43 @@ function MessageContent_({
       <AIElementMessageContent className={className}>
         {filesList}
         <div data-reference-batch-id={message.id} className="contents">
+          {reasoningToDisplay && !multiRoundTrace && (
+            <Reasoning isStreaming={isLoading && !contentToDisplay}>
+              <ReasoningTrigger />
+              <ReasoningContent>
+                {isSingleRoundRag || displaySources.length > 0 ? (
+                  <MarkdownContent
+                    content={reasoningToDisplay}
+                    isLoading={isLoading}
+                    rehypePlugins={markdownRehypePlugins}
+                    components={components}
+                    className="text-muted-foreground text-sm"
+                  />
+                ) : (
+                  reasoningToDisplay
+                )}
+              </ReasoningContent>
+            </Reasoning>
+          )}
           {multiRoundTrace ? (
             <MultiRoundRagContent
               trace={multiRoundTrace}
               content={contentToDisplay}
               displaySources={displaySources}
               isLoading={isLoading}
-              rehypePlugins={[...rehypePlugins, [rehypeKatex, { output: 'html' }]]}
+              rehypePlugins={markdownRehypePlugins}
               components={components}
             />
-          ) : (
+          ) : contentToDisplay ? (
             <AssistantDisplayContent
               content={contentToDisplay}
               displaySources={displaySources}
               isLoading={isLoading}
               renderSkillDisplayBlocks={false}
-              rehypePlugins={[...rehypePlugins, [rehypeKatex, { output: 'html' }]]}
+              rehypePlugins={markdownRehypePlugins}
               components={components}
             />
-          )}
+          ) : null}
         </div>
         {(isSingleRoundRag || isMultiRoundRag) && displaySources.length > 0 && (
           <FrontendRagSourcesPanel
@@ -840,7 +852,7 @@ function FrontendRagSourcesPanel({
   const [open, setOpen] = useState(false);
   if (sources.length === 0) return null;
   return (
-    <Collapsible open={open} onOpenChange={setOpen} className="mt-3 w-full max-w-5xl">
+    <Collapsible open={open} onOpenChange={setOpen} className="mt-3 w-full max-w-5xl">
       <CollapsibleTrigger asChild>
         <button
           type="button"
@@ -946,17 +958,17 @@ function MultiRoundRagContent({
   isLoading: boolean;
   rehypePlugins: MarkdownContentProps['rehypePlugins'];
   components: MarkdownContentProps['components'];
-}) {
-  const [open, setOpen] = useState(!trace.isComplete);
-  useEffect(() => {
-    // Keep the active turn expanded so each new retrieval is visible as it
-    // arrives, then match GeneralQA by collapsing the completed trace.  The
-    // header still carries the total round count for quick scanning.
-    setOpen(!trace.isComplete);
-  }, [trace.isComplete]);
+}) {
+  const [open, setOpen] = useState(!trace.isComplete);
+  useEffect(() => {
+    // Keep the active turn expanded so each new retrieval is visible as it
+    // arrives, then match GeneralQA by collapsing the completed trace.  The
+    // header still carries the total round count for quick scanning.
+    setOpen(!trace.isComplete);
+  }, [trace.isComplete]);
 
   return (
-    <div className="w-full min-w-0 max-w-5xl space-y-4 overflow-hidden">
+    <div className="w-full min-w-0 max-w-5xl space-y-4 overflow-hidden">
       <section className="border-border/70 overflow-hidden rounded-xl border bg-muted/20">
         <button
           type="button"
@@ -970,42 +982,43 @@ function MultiRoundRagContent({
               <Loader2Icon className="text-primary size-4 shrink-0 animate-spin" />
             )}
             <div>
-              <div className="text-sm font-medium">多轮检索思考过程</div>
-              <div className="text-muted-foreground text-xs">
-                {trace.isComplete
-                  ? `共 ${trace.rounds.length} 轮推理与检索，可展开查看细节`
-                  : '正在分析问题并检索资料'}
-              </div>
+              <div className="text-sm font-medium">多轮检索思考过程</div>
+              <div className="text-muted-foreground text-xs">
+                {trace.isComplete
+                  ? `共 ${trace.rounds.length} 轮推理与检索，可展开查看细节`
+                  : '正在分析问题并检索资料'}
+              </div>
             </div>
           </div>
           <ChevronDownIcon
             className={cn('text-muted-foreground size-4 transition-transform', !open && '-rotate-90')}
           />
         </button>
-        {open && (
-          <div className="border-border/60 space-y-3 border-t px-4 py-3">
-            {trace.rounds.map((round, index) => (
-              <div
-                key={round.id}
-                className={cn(
-                  'relative pl-5',
-                  index < trace.rounds.length - 1 && 'pb-1',
-                )}
-              >
-                {index < trace.rounds.length - 1 && (
-                  <span className="bg-border absolute bottom-0 left-[7px] top-8 w-px" />
-                )}
-                <span className="border-primary bg-background absolute left-0 top-4 size-3.5 rounded-full border-2" />
-                <MultiRoundRagRoundCard
-                  index={index + 1}
-                  round={round}
-                  isLoading={isLoading}
-                  rehypePlugins={rehypePlugins}
-                  components={components}
-                />
-              </div>
-            ))}
-          </div>
+        {open && (
+          <div className="border-border/60 space-y-3 border-t px-4 py-3">
+            {trace.rounds.map((round, index) => (
+              <div
+                key={round.id}
+                className={cn(
+                  'relative pl-5',
+                  index < trace.rounds.length - 1 && 'pb-1',
+                )}
+              >
+                {index < trace.rounds.length - 1 && (
+                  <span className="bg-border absolute bottom-0 left-[7px] top-8 w-px" />
+                )}
+                <span className="border-primary bg-background absolute left-0 top-4 size-3.5 rounded-full border-2" />
+                <MultiRoundRagRoundCard
+                  index={index + 1}
+                  round={round}
+                  displaySources={displaySources}
+                  isLoading={isLoading}
+                  rehypePlugins={rehypePlugins}
+                  components={components}
+                />
+              </div>
+            ))}
+          </div>
         )}
       </section>
       <section className="border-border/70 rounded-xl border p-4">
@@ -1030,240 +1043,249 @@ function MultiRoundRagContent({
   );
 }
 
-function MultiRoundRagRoundCard({
+function MultiRoundRagRoundCard({
   index,
   round,
+  displaySources,
   isLoading,
   rehypePlugins,
   components,
 }: {
   index: number;
   round: MultiRoundRagRound;
+  displaySources: RuntimeDisplaySource[];
   isLoading: boolean;
   rehypePlugins: MarkdownContentProps['rehypePlugins'];
   components: MarkdownContentProps['components'];
-}) {
-  const [open, setOpen] = useState(!round.query || round.isSearching);
-  useEffect(() => {
-    // A newly created round stays expanded to expose its thought and loading
-    // query. Once that query has returned, retain only the round header until
-    // the user asks to inspect the completed detail again.
-    setOpen(!round.query || round.isSearching);
-  }, [round.id, round.isSearching, round.query]);
-
-  return (
-    <Collapsible open={open} onOpenChange={setOpen}>
-      <article className="border-border/60 bg-background/80 overflow-hidden rounded-xl border shadow-sm">
-        <CollapsibleTrigger asChild>
-          <button
-            type="button"
-            className="hover:bg-muted/40 focus-visible:ring-ring flex min-h-11 w-full min-w-0 items-center gap-2 px-3 py-2.5 text-left text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-inset"
-          >
-            <span className="bg-primary/10 text-primary flex size-6 shrink-0 items-center justify-center rounded-full text-xs font-semibold">
-              {index}
-            </span>
-            <span className="min-w-0 flex-1 truncate">{round.title}</span>
-            {round.isSearching ? (
-              <span className="bg-primary/10 text-primary inline-flex shrink-0 items-center gap-1 rounded-full px-2 py-1 text-[11px] font-medium">
-                <Loader2Icon className="size-3 motion-reduce:animate-none animate-spin" />
-                检索中
-              </span>
-            ) : (
-              <span className="text-muted-foreground shrink-0 text-[11px]">已完成</span>
-            )}
-            <ChevronDownIcon
-              className={cn(
-                'text-muted-foreground size-4 shrink-0 transition-transform duration-200 motion-reduce:transition-none',
-                !open && '-rotate-90',
-              )}
-            />
-          </button>
-        </CollapsibleTrigger>
-        <CollapsibleContent className="border-border/50 border-t px-3 pb-3">
-          <div className="pt-3">
-            {round.thought && (
-              <MultiRoundRagDetailSection
-                title="思考内容"
-                icon={<SparklesIcon className="text-primary size-4" />}
-              >
-                <MarkdownContent
-                  content={round.thought}
-                  isLoading={isLoading}
-                  rehypePlugins={rehypePlugins}
-                  components={components}
-                  className="text-muted-foreground text-sm"
-                />
-              </MultiRoundRagDetailSection>
-            )}
-            {round.body && (
-              <section className="border-border/50 mt-3 rounded-lg border p-3">
-                <div className="text-muted-foreground mb-2 text-xs font-medium">本轮正文</div>
-                <MarkdownContent
-                  content={round.body}
-                  isLoading={isLoading}
-                  rehypePlugins={rehypePlugins}
-                  components={components}
-                  className="text-sm"
-                />
-              </section>
-            )}
-            {(round.query || round.isSearching) && (
-              <div
-                className={cn(
-                  'border-primary/20 bg-primary/5 text-primary mt-3 inline-flex max-w-full items-center gap-1.5 rounded-full border px-2.5 py-1.5 text-xs',
-                  round.isSearching && 'animate-pulse motion-reduce:animate-none',
-                )}
-                aria-live={round.isSearching ? 'polite' : undefined}
-              >
-                {round.isSearching ? (
-                  <Loader2Icon className="size-3.5 shrink-0 motion-reduce:animate-none animate-spin" />
-                ) : (
-                  <SearchIcon className="size-3.5 shrink-0" />
-                )}
-                <span className="shrink-0 text-muted-foreground">检索词</span>
-                <span className="min-w-0 truncate font-medium">
-                  {round.query || '正在生成检索词…'}
-                </span>
-              </div>
-            )}
-            {(round.query || round.isSearching || round.retrievals.length > 0) && (
-              <MultiRoundRagDetailSection
-                defaultOpen={!round.isSearching}
-                title="检索内容"
-                icon={<FileIcon className="text-primary size-3.5" />}
-                suffix={<span className="text-muted-foreground text-xs">{round.retrievals.length} 条</span>}
-              >
-                {round.isSearching ? (
-                  <div className="text-muted-foreground flex items-center gap-2 py-1 text-sm">
-                    <Loader2Icon className="text-primary size-4 motion-reduce:animate-none animate-spin" />
-                    正在召回相关内容…
-                  </div>
-                ) : (
-                  <div className="space-y-2">
-                    {round.retrievals.length > 0 ? (
-                      round.retrievals.map((item, resultIndex) => (
-                        <MultiRoundRagResultCard
-                          key={`${item.id ?? item.title}-${resultIndex}`}
-                          item={item}
-                          index={resultIndex + 1}
-                        />
-                      ))
-                    ) : (
-                      <p className="text-muted-foreground py-1 text-sm">
-                        本轮未召回相关资料。
-                      </p>
-                    )}
-                  </div>
-                )}
-              </MultiRoundRagDetailSection>
-            )}
-          </div>
-        </CollapsibleContent>
-      </article>
-    </Collapsible>
-  );
-}
-
-function MultiRoundRagDetailSection({
-  title,
-  icon,
-  suffix,
-  defaultOpen = true,
-  children,
-}: {
-  title: string;
-  icon: ReactNode;
-  suffix?: ReactNode;
-  defaultOpen?: boolean;
-  children: ReactNode;
-}) {
-  const [open, setOpen] = useState(defaultOpen);
-  useEffect(() => setOpen(defaultOpen), [defaultOpen]);
-
-  return (
-    <Collapsible open={open} onOpenChange={setOpen}>
-      <section className="border-border/50 bg-muted/20 mt-3 overflow-hidden rounded-lg border">
-        <CollapsibleTrigger asChild>
-          <button
-            type="button"
-            className="hover:bg-muted/40 focus-visible:ring-ring flex min-h-11 w-full min-w-0 items-center gap-2 px-3 py-2 text-left transition-colors focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-inset"
-          >
-            {icon}
-            <span className="min-w-0 flex-1 text-sm font-medium">{title}</span>
-            {suffix}
-            <ChevronDownIcon
-              className={cn(
-                'text-muted-foreground size-4 shrink-0 transition-transform duration-200 motion-reduce:transition-none',
-                !open && '-rotate-90',
-              )}
-            />
-          </button>
-        </CollapsibleTrigger>
-        <CollapsibleContent className="border-border/50 border-t px-3 pb-3 pt-2.5">
-          {children}
-        </CollapsibleContent>
-      </section>
-    </Collapsible>
-  );
-}
-
-function MultiRoundRagResultCard({
-  item,
-  index,
-}: {
-  item: MultiRoundRagRound['retrievals'][number];
-  index: number;
-}) {
-  const [open, setOpen] = useState(index === 1);
-  const sourceMeta = [item.source, item.time].filter(Boolean).join(' · ');
-
-  return (
-    <Collapsible open={open} onOpenChange={setOpen}>
-      <div className="border-border/60 hover:border-primary/30 bg-background rounded-lg border transition-colors">
-        <CollapsibleTrigger asChild>
-          <button
-            type="button"
-            className="hover:bg-muted/40 focus-visible:ring-ring flex min-h-11 w-full min-w-0 items-center gap-2 rounded-lg px-3 py-2.5 text-left transition-colors focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-inset"
-          >
-            <span className="bg-primary/10 text-primary flex size-5 shrink-0 items-center justify-center rounded-md text-[11px] font-semibold">
-              {index}
-            </span>
-            <span className="min-w-0 flex-1 truncate text-sm font-medium" title={item.title}>
-              {item.title}
-            </span>
-            <ChevronDownIcon
-              className={cn(
-                'text-muted-foreground size-4 shrink-0 transition-transform duration-200 motion-reduce:transition-none',
-                !open && '-rotate-90',
-              )}
-            />
-          </button>
-        </CollapsibleTrigger>
-        <CollapsibleContent className="border-border/50 border-t px-3 pb-3 pt-2.5">
-          <p className="text-muted-foreground text-sm leading-6 whitespace-pre-wrap">
-            {item.content || '暂无资料摘要'}
-          </p>
-          {(sourceMeta || item.url) && (
-            <div className="text-muted-foreground mt-2 flex flex-wrap items-center gap-x-2 gap-y-1 text-xs">
-              {sourceMeta && <span>{sourceMeta}</span>}
-              {item.url && (
-                <a
-                  href={item.url}
-                  target="_blank"
-                  rel="noopener noreferrer"
-                  className="text-primary hover:underline inline-flex items-center gap-1"
-                >
-                  查看原文
-                  <ExternalLinkIcon className="size-3" />
-                </a>
-              )}
-            </div>
-          )}
-        </CollapsibleContent>
-      </div>
-    </Collapsible>
-  );
-}
+}) {
+  const [open, setOpen] = useState(!round.query || round.isSearching);
+  useEffect(() => {
+    // A newly created round stays expanded to expose its thought and loading
+    // query. Once that query has returned, retain only the round header until
+    // the user asks to inspect the completed detail again.
+    setOpen(!round.query || round.isSearching);
+  }, [round.id, round.isSearching, round.query]);
+  const thoughtContent = useMemo(
+    () =>
+      applyInlineCitations(round.thought, displaySources, {
+        appendSummary: false,
+      }),
+    [displaySources, round.thought],
+  );
+
+  return (
+    <Collapsible open={open} onOpenChange={setOpen}>
+      <article className="border-border/60 bg-background/80 overflow-hidden rounded-xl border shadow-sm">
+        <CollapsibleTrigger asChild>
+          <button
+            type="button"
+            className="hover:bg-muted/40 focus-visible:ring-ring flex min-h-11 w-full min-w-0 items-center gap-2 px-3 py-2.5 text-left text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-inset"
+          >
+            <span className="bg-primary/10 text-primary flex size-6 shrink-0 items-center justify-center rounded-full text-xs font-semibold">
+              {index}
+            </span>
+            <span className="min-w-0 flex-1 truncate">{round.title}</span>
+            {round.isSearching ? (
+              <span className="bg-primary/10 text-primary inline-flex shrink-0 items-center gap-1 rounded-full px-2 py-1 text-[11px] font-medium">
+                <Loader2Icon className="size-3 motion-reduce:animate-none animate-spin" />
+                检索中
+              </span>
+            ) : (
+              <span className="text-muted-foreground shrink-0 text-[11px]">已完成</span>
+            )}
+            <ChevronDownIcon
+              className={cn(
+                'text-muted-foreground size-4 shrink-0 transition-transform duration-200 motion-reduce:transition-none',
+                !open && '-rotate-90',
+              )}
+            />
+          </button>
+        </CollapsibleTrigger>
+        <CollapsibleContent className="border-border/50 border-t px-3 pb-3">
+          <div className="pt-3">
+            {round.thought && (
+              <MultiRoundRagDetailSection
+                title="思考内容"
+                icon={<SparklesIcon className="text-primary size-4" />}
+              >
+                <MarkdownContent
+                  content={thoughtContent}
+                  isLoading={isLoading}
+                  rehypePlugins={rehypePlugins}
+                  components={components}
+                  className="text-muted-foreground text-sm"
+                />
+              </MultiRoundRagDetailSection>
+            )}
+            {round.body && (
+              <section className="border-border/50 mt-3 rounded-lg border p-3">
+                <div className="text-muted-foreground mb-2 text-xs font-medium">本轮正文</div>
+                <MarkdownContent
+                  content={round.body}
+                  isLoading={isLoading}
+                  rehypePlugins={rehypePlugins}
+                  components={components}
+                  className="text-sm"
+                />
+              </section>
+            )}
+            {(round.query || round.isSearching) && (
+              <div
+                className={cn(
+                  'border-primary/20 bg-primary/5 text-primary mt-3 inline-flex max-w-full items-center gap-1.5 rounded-full border px-2.5 py-1.5 text-xs',
+                  round.isSearching && 'animate-pulse motion-reduce:animate-none',
+                )}
+                aria-live={round.isSearching ? 'polite' : undefined}
+              >
+                {round.isSearching ? (
+                  <Loader2Icon className="size-3.5 shrink-0 motion-reduce:animate-none animate-spin" />
+                ) : (
+                  <SearchIcon className="size-3.5 shrink-0" />
+                )}
+                <span className="shrink-0 text-muted-foreground">检索词</span>
+                <span className="min-w-0 truncate font-medium">
+                  {round.query || '正在生成检索词…'}
+                </span>
+              </div>
+            )}
+            {(round.query || round.isSearching || round.retrievals.length > 0) && (
+              <MultiRoundRagDetailSection
+                defaultOpen={!round.isSearching}
+                title="检索内容"
+                icon={<FileIcon className="text-primary size-3.5" />}
+                suffix={<span className="text-muted-foreground text-xs">{round.retrievals.length} 条</span>}
+              >
+                {round.isSearching ? (
+                  <div className="text-muted-foreground flex items-center gap-2 py-1 text-sm">
+                    <Loader2Icon className="text-primary size-4 motion-reduce:animate-none animate-spin" />
+                    正在召回相关内容…
+                  </div>
+                ) : (
+                  <div className="space-y-2">
+                    {round.retrievals.length > 0 ? (
+                      round.retrievals.map((item, resultIndex) => (
+                        <MultiRoundRagResultCard
+                          key={`${item.id ?? item.title}-${resultIndex}`}
+                          item={item}
+                          index={resultIndex + 1}
+                        />
+                      ))
+                    ) : (
+                      <p className="text-muted-foreground py-1 text-sm">
+                        本轮未召回相关资料。
+                      </p>
+                    )}
+                  </div>
+                )}
+              </MultiRoundRagDetailSection>
+            )}
+          </div>
+        </CollapsibleContent>
+      </article>
+    </Collapsible>
+  );
+}
+
+function MultiRoundRagDetailSection({
+  title,
+  icon,
+  suffix,
+  defaultOpen = true,
+  children,
+}: {
+  title: string;
+  icon: ReactNode;
+  suffix?: ReactNode;
+  defaultOpen?: boolean;
+  children: ReactNode;
+}) {
+  const [open, setOpen] = useState(defaultOpen);
+  useEffect(() => setOpen(defaultOpen), [defaultOpen]);
+
+  return (
+    <Collapsible open={open} onOpenChange={setOpen}>
+      <section className="border-border/50 bg-muted/20 mt-3 overflow-hidden rounded-lg border">
+        <CollapsibleTrigger asChild>
+          <button
+            type="button"
+            className="hover:bg-muted/40 focus-visible:ring-ring flex min-h-11 w-full min-w-0 items-center gap-2 px-3 py-2 text-left transition-colors focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-inset"
+          >
+            {icon}
+            <span className="min-w-0 flex-1 text-sm font-medium">{title}</span>
+            {suffix}
+            <ChevronDownIcon
+              className={cn(
+                'text-muted-foreground size-4 shrink-0 transition-transform duration-200 motion-reduce:transition-none',
+                !open && '-rotate-90',
+              )}
+            />
+          </button>
+        </CollapsibleTrigger>
+        <CollapsibleContent className="border-border/50 border-t px-3 pb-3 pt-2.5">
+          {children}
+        </CollapsibleContent>
+      </section>
+    </Collapsible>
+  );
+}
+
+function MultiRoundRagResultCard({
+  item,
+  index,
+}: {
+  item: MultiRoundRagRound['retrievals'][number];
+  index: number;
+}) {
+  const [open, setOpen] = useState(index === 1);
+  const sourceMeta = [item.source, item.time].filter(Boolean).join(' · ');
+
+  return (
+    <Collapsible open={open} onOpenChange={setOpen}>
+      <div className="border-border/60 hover:border-primary/30 bg-background rounded-lg border transition-colors">
+        <CollapsibleTrigger asChild>
+          <button
+            type="button"
+            className="hover:bg-muted/40 focus-visible:ring-ring flex min-h-11 w-full min-w-0 items-center gap-2 rounded-lg px-3 py-2.5 text-left transition-colors focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-inset"
+          >
+            <span className="bg-primary/10 text-primary flex size-5 shrink-0 items-center justify-center rounded-md text-[11px] font-semibold">
+              {index}
+            </span>
+            <span className="min-w-0 flex-1 truncate text-sm font-medium" title={item.title}>
+              {item.title}
+            </span>
+            <ChevronDownIcon
+              className={cn(
+                'text-muted-foreground size-4 shrink-0 transition-transform duration-200 motion-reduce:transition-none',
+                !open && '-rotate-90',
+              )}
+            />
+          </button>
+        </CollapsibleTrigger>
+        <CollapsibleContent className="border-border/50 border-t px-3 pb-3 pt-2.5">
+          <p className="text-muted-foreground text-sm leading-6 whitespace-pre-wrap">
+            {item.content || '暂无资料摘要'}
+          </p>
+          {(sourceMeta || item.url) && (
+            <div className="text-muted-foreground mt-2 flex flex-wrap items-center gap-x-2 gap-y-1 text-xs">
+              {sourceMeta && <span>{sourceMeta}</span>}
+              {item.url && (
+                <a
+                  href={item.url}
+                  target="_blank"
+                  rel="noopener noreferrer"
+                  className="text-primary hover:underline inline-flex items-center gap-1"
+                >
+                  查看原文
+                  <ExternalLinkIcon className="size-3" />
+                </a>
+              )}
+            </div>
+          )}
+        </CollapsibleContent>
+      </div>
+    </Collapsible>
+  );
+}
 
 function resolveSingleRoundRagSourceUrl(
   source: RuntimeDisplaySource | null,
diff --git a/frontend-web/src/components/workspace/workspace-nav-menu.tsx b/frontend-web/src/components/workspace/workspace-nav-menu.tsx
index 25865950..d8dd23c6 100644
--- a/frontend-web/src/components/workspace/workspace-nav-menu.tsx
+++ b/frontend-web/src/components/workspace/workspace-nav-menu.tsx
@@ -10,6 +10,7 @@ import {
   DatabaseZapIcon,
   GaugeIcon,
   HistoryIcon,
+  ListTreeIcon,
   MessageCircleQuestionIcon,
   ScrollTextIcon,
   Settings2Icon,
@@ -226,6 +227,16 @@ export function WorkspaceNavMenu() {
                       技能管理
                     </DropdownMenuItem>
                   )}
+                  {/* 与全量模式左下角「设置和更多」下拉里的菜单管理同入口（管理员）：
+                      简洁模式没有全局侧边栏，这里放一份保证两种模式都能进菜单管理。 */}
+                  {isAdmin && (
+                    <DropdownMenuItem
+                      onClick={() => navigate("/page/strategy/admin/menu-management")}
+                    >
+                      <ListTreeIcon />
+                      菜单管理
+                    </DropdownMenuItem>
+                  )}
                   <DropdownMenuItem
                     onClick={() => navigate("/page/workspace/admin/tags")}
                   >
diff --git a/frontend-web/src/core/agents/types.ts b/frontend-web/src/core/agents/types.ts
index 26bd7e33..06718d76 100644
--- a/frontend-web/src/core/agents/types.ts
+++ b/frontend-web/src/core/agents/types.ts
@@ -61,9 +61,11 @@ export interface Agent {
   /** 会商席位「技能识别强化」开关。作为会商席位运行时，是否把本智能体配置的技能
    * (名称 + SKILL.md 路径 + 必须先 read_file 的硬指令) 拼进本轮任务以补偿弱模型。
    * 与业务链条同名开关取 OR。null/false = 关 (默认)。 */
-  seat_skill_directive?: boolean | null;
-  /** DeerFlow LLMWiki mapping ids this agent allows conversations to select. */
-  llmwiki_knowledge_base_ids?: string[] | null;
+  seat_skill_directive?: boolean | null;
+  /** 岗位职责。会商智能体配置面板展示在常用问题上方。空/缺省 = 未填写。 */
+  job_responsibilities?: string | null;
+  /** DeerFlow LLMWiki mapping ids this agent allows conversations to select. */
+  llmwiki_knowledge_base_ids?: string[] | null;
   tags?: Tag[];
   featured_order?: number | null;
   /** Whether the current user has favorited (收藏) this agent. */
@@ -98,8 +100,10 @@ export interface CreateAgentRequest {
   common_questions?: AgentQuestion[] | null;
   template_name?: string | null;
   /** 会商席位「技能识别强化」开关；缺省/false = 关。 */
-  seat_skill_directive?: boolean | null;
-  llmwiki_knowledge_base_ids?: string[] | null;
+  seat_skill_directive?: boolean | null;
+  /** 岗位职责；缺省/空 = 未填写。 */
+  job_responsibilities?: string | null;
+  llmwiki_knowledge_base_ids?: string[] | null;
 }
 
 export interface UpdateAgentRequest {
@@ -116,6 +120,8 @@ export interface UpdateAgentRequest {
   /** 入库模板名；传空字符串表示清除。 */
   template_name?: string | null;
   /** 会商席位「技能识别强化」开关；缺省/false = 关。 */
-  seat_skill_directive?: boolean | null;
-  llmwiki_knowledge_base_ids?: string[] | null;
-}
+  seat_skill_directive?: boolean | null;
+  /** 岗位职责；传空字符串表示清除。 */
+  job_responsibilities?: string | null;
+  llmwiki_knowledge_base_ids?: string[] | null;
+}
diff --git a/frontend-web/src/core/skill-display/inline-citations.test.ts b/frontend-web/src/core/skill-display/inline-citations.test.ts
index 0b4f3cc1..fb3eb03c 100644
--- a/frontend-web/src/core/skill-display/inline-citations.test.ts
+++ b/frontend-web/src/core/skill-display/inline-citations.test.ts
@@ -82,3 +82,14 @@ test("accepts spaced and full-width square brackets", () => {
   assert.equal(out.includes(citation(1)), true);
   assert.equal(out.includes(citation(2)), true);
 });
+
+test("can skip the trailing source summary", () => {
+  const withSummary = applyInlineCitations("没有编号。", makeSources(2));
+  assert.equal(withSummary.includes("参考来源"), true);
+
+  const withoutSummary = applyInlineCitations("没有编号。", makeSources(2), {
+    appendSummary: false,
+  });
+  assert.equal(withoutSummary, "没有编号。");
+  assert.equal(withoutSummary.includes("#reference-"), false);
+});
diff --git a/frontend-web/src/core/skill-display/runtime.ts b/frontend-web/src/core/skill-display/runtime.ts
index 970c7433..fbc63213 100644
--- a/frontend-web/src/core/skill-display/runtime.ts
+++ b/frontend-web/src/core/skill-display/runtime.ts
@@ -509,6 +509,7 @@ export function normalizeDefaultDisplaySources(payload: unknown): DisplaySourceI
 export function applyInlineCitations(
   content: string,
   sources: RuntimeDisplaySource[],
+  options?: { appendSummary?: boolean },
 ): string {
   if (!content || sources.length === 0) {
     return content;
@@ -563,6 +564,9 @@ export function applyInlineCitations(
   if (replaced || hasCitationMarkdown(next)) {
     return next;
   }
+  if (options?.appendSummary === false) {
+    return next;
+  }
   return appendCitationSummary(next, referenceSources);
 }
 
diff --git a/frontend-web/src/eightbf-roundtable/components/EightBfStep1Workspace.tsx b/frontend-web/src/eightbf-roundtable/components/EightBfStep1Workspace.tsx
index 2af55a28..5d4423a0 100644
--- a/frontend-web/src/eightbf-roundtable/components/EightBfStep1Workspace.tsx
+++ b/frontend-web/src/eightbf-roundtable/components/EightBfStep1Workspace.tsx
@@ -1,4 +1,5 @@
 import { useEffect, useMemo, useState, type ReactNode } from "react";
+import { useSearchParams } from "react-router-dom";
 import {
   ChevronDown,
   ChevronLeft,
@@ -19,15 +20,23 @@ import backArrowIcon from "@/assets/eightbf-roundtable/back-arrow.svg";
 import agentConfigLogo from "@/assets/position-roundtable/agent-config-logo.svg";
 import collapsePanelIcon from "@/assets/position-roundtable/collapse-panel.svg";
 import doubleArrowDecoration from "@/assets/position-roundtable/double-arrow-decoration.svg";
-import { getEightBfRoundtableBackUrl } from "@/eightbf-roundtable/api/task-context";
+import {
+  parentReturnRoute,
+  requestEightBfParentReturn,
+} from "@/eightbf-roundtable/lib/parent-return";
 import { useEightBfAnalysisResults } from "@/eightbf-roundtable/lib/collect-md-artifacts";
 import { PositionIntentInput } from "@/position-roundtable/components/PositionIntentInput";
+import {
+  positionNameById,
+  syncPositionOptionsFromRoles,
+} from "@/position-roundtable/lib/positions";
 import type { DraftStep3SummarySnapshot } from "@/roundtable-planning/api/drafts";
 import type { ThreadIdMap } from "@/roundtable-planning/api/multi-agent";
 import "@/position-roundtable/styles/intent-understanding.css";
 import type { Step1PanelProps } from "@/roundtable-planning/components/Step1Panel";
 import { Step1Panel } from "@/roundtable-planning/components/Step1Panel";
 import type { BusinessChain, ChainSeat } from "@/roundtable-planning/lib/business-chain";
+import { usePositionRoles } from "@/strategy-components/api/position-roles";
 import { cn } from "@/lib/utils";
 
 type IntentRecord = Record<string, unknown>;
@@ -47,13 +56,16 @@ export interface EightBfStep1PanelProps extends Step1PanelProps {
 }
 
 export function EightBfRoundtableBackButton() {
+  const [searchParams] = useSearchParams();
   const handleBack = () => {
-    const url = getEightBfRoundtableBackUrl();
-    if (!url) {
-      toast.warning("尚未配置返回地址，请在 runtime-config.js 填写 VITE_8BF_ROUNDTABLE_BACK_URL");
-      return;
-    }
-    window.location.href = url;
+    const route = parentReturnRoute(
+      searchParams.get("parentReturnTo")
+        ?? searchParams.get("parentRoute")
+        ?? searchParams.get("returnTo")
+        ?? "/list",
+    );
+    if (requestEightBfParentReturn(route)) return;
+    toast.warning("尚未配置返回地址，请在 runtime-config.js 填写 VITE_8BF_ROUNDTABLE_BACK_URL");
   };
   return (
     <button
@@ -301,6 +313,10 @@ function EightBfAgentWorkspace({
   step3Summary: DraftStep3SummarySnapshot | null;
 }) {
   const { agents, isLoading } = useAgents();
+  const { data: positionRoles = [] } = usePositionRoles();
+  useEffect(() => {
+    syncPositionOptionsFromRoles(positionRoles);
+  }, [positionRoles]);
   const seats = chain?.seats ?? [];
   const [selectedAgentId, setSelectedAgentId] = useState<string | null>(null);
   const [agentKeyword, setAgentKeyword] = useState("");
@@ -400,7 +416,7 @@ function EightBfAgentWorkspace({
               onAskQuestion={() => toast.info("请在中间对话区向任务理解智能体发起问题")}
               showRecentChats={false}
               avatarSrc={agentConfigLogo}
-              positionName={chain?.title || "8BF 业务链"}
+              positionName={positionNameById(selectedSeat?.position_id)}
               skillsIcon={<img src={doubleArrowDecoration} alt="" aria-hidden="true" />}
               skillsTitle="技能列表"
               showCommonQuestions
diff --git a/frontend-web/src/eightbf-roundtable/lib/parent-return.ts b/frontend-web/src/eightbf-roundtable/lib/parent-return.ts
new file mode 100644
index 00000000..bb60e43d
--- /dev/null
+++ b/frontend-web/src/eightbf-roundtable/lib/parent-return.ts
@@ -0,0 +1,128 @@
+import { useEffect } from "react";
+
+import { emitWujieEvent, isWujieEnv } from "@/core/embed/wujie";
+import { getEightBfRoundtableBackUrl } from "@/eightbf-roundtable/api/task-context";
+
+export const EIGHTBF_PARENT_RETURN_SOURCE = "deerflow-8bf-roundtable";
+export const EIGHTBF_PARENT_RETURN_TYPE = "deerflow:8bf-roundtable:return";
+export const EIGHTBF_BRIDGE_HOST_SOURCE = "code-zk-roundtable-host";
+export const EIGHTBF_BRIDGE_INIT_TYPE = "code-zk:8bf-roundtable-bridge:init";
+export const EIGHTBF_BRIDGE_READY_TYPE = "deerflow:8bf-roundtable:bridge-ready";
+
+export interface EightBfParentReturnMessage {
+  source: typeof EIGHTBF_PARENT_RETURN_SOURCE;
+  type: typeof EIGHTBF_PARENT_RETURN_TYPE;
+  payload: {
+    route: string | null;
+    fallback: "history-back";
+  };
+}
+
+type EightBfBridgeInitMessage = {
+  source: typeof EIGHTBF_BRIDGE_HOST_SOURCE;
+  type: typeof EIGHTBF_BRIDGE_INIT_TYPE;
+};
+
+/**
+ * `parentReturnTo` belongs to the embedding application. Keep it a route/hash
+ * path so same-origin fallback cannot become an open redirect.
+ */
+export function parentReturnRoute(value: string | null | undefined): string | null {
+  const route = value?.trim();
+  if (!route || route.length > 2048) return null;
+  if ((!route.startsWith("/") && !route.startsWith("#")) || route.startsWith("//") || route.includes("\\")) {
+    return null;
+  }
+  return route;
+}
+
+function isInIframe(): boolean {
+  if (typeof window === "undefined") return false;
+  try {
+    return window.parent !== window;
+  } catch {
+    return true;
+  }
+}
+
+const embeddingBridgePort = { current: null as MessagePort | null };
+
+/**
+ * Accept a MessagePort from code-zk. Wujie can lift this iframe into the
+ * outer host document, so `window.parent` is no longer code-zk. The port
+ * still talks to the Vue page that created the channel.
+ */
+export function useEightBfParentReturnBridge(): void {
+  useEffect(() => {
+    const receiveBridgePort = (event: MessageEvent<EightBfBridgeInitMessage>) => {
+      const message = event.data;
+      const bridgePort = event.ports?.[0];
+      if (
+        message?.source !== EIGHTBF_BRIDGE_HOST_SOURCE
+        || message?.type !== EIGHTBF_BRIDGE_INIT_TYPE
+        || !bridgePort
+      ) {
+        return;
+      }
+      embeddingBridgePort.current?.close();
+      embeddingBridgePort.current = bridgePort;
+      bridgePort.start?.();
+      bridgePort.postMessage({
+        source: EIGHTBF_PARENT_RETURN_SOURCE,
+        type: EIGHTBF_BRIDGE_READY_TYPE,
+      });
+    };
+
+    window.addEventListener("message", receiveBridgePort);
+    return () => {
+      window.removeEventListener("message", receiveBridgePort);
+      embeddingBridgePort.current?.close();
+      embeddingBridgePort.current = null;
+    };
+  }, []);
+}
+
+/**
+ * Ask the embedding Vue host to leave this iframe.
+ *
+ * Topology: Wujie host → code-zk → DeerFlow iframe. Wujie may reparent the
+ * iframe so native `window.parent.postMessage` never reaches code-zk.
+ * MessageChannel is the channel that survives that. Native postMessage is
+ * only the fallback for a plain iframe.
+ */
+export function requestEightBfParentReturn(route: string | null): boolean {
+  const message: EightBfParentReturnMessage = {
+    source: EIGHTBF_PARENT_RETURN_SOURCE,
+    type: EIGHTBF_PARENT_RETURN_TYPE,
+    payload: { route, fallback: "history-back" },
+  };
+
+  if (typeof window === "undefined") return false;
+
+  const bridgePort = embeddingBridgePort.current;
+  if (bridgePort) {
+    try {
+      bridgePort.postMessage(message);
+      return true;
+    } catch {
+      // Host unmounted/reloaded. Fall through to native bridges.
+    }
+  }
+
+  if (isInIframe()) {
+    window.parent.postMessage(message, "*");
+    return true;
+  }
+
+  if (isWujieEnv()) {
+    emitWujieEvent("8bf-roundtable:return", message.payload);
+    return true;
+  }
+
+  const url = getEightBfRoundtableBackUrl().trim();
+  if (url) {
+    window.location.href = url;
+    return true;
+  }
+  return false;
+}
diff --git a/frontend-web/src/eightbf-roundtable/pages/EightBfRoundtablePage.tsx b/frontend-web/src/eightbf-roundtable/pages/EightBfRoundtablePage.tsx
index 870fa65e..582b0a39 100644
--- a/frontend-web/src/eightbf-roundtable/pages/EightBfRoundtablePage.tsx
+++ b/frontend-web/src/eightbf-roundtable/pages/EightBfRoundtablePage.tsx
@@ -25,6 +25,7 @@ import { PositionBusinessChainFlow } from "@/position-roundtable/components/Posi
 import RoundtablePlanningPage from "@/roundtable-planning/pages/RoundtablePlanningPage";
 import { listChains } from "@/roundtable-planning/api/chains";
 import type { BusinessChain } from "@/roundtable-planning/lib/business-chain";
+import { useEightBfParentReturnBridge } from "@/eightbf-roundtable/lib/parent-return";
 import {
   listBusinessMappings,
   updateBusinessMapping,
@@ -44,6 +45,7 @@ const BUSINESS_CODE = "8BF";
  * 不引用岗位会商的页面、会话或接口。
  */
 export default function EightBfRoundtablePage({ compactChrome = false }: { compactChrome?: boolean }) {
+  useEightBfParentReturnBridge();
   const auth = getStoredAuth();
   const [searchParams] = useSearchParams();
   const taskId = useMemo(
diff --git a/frontend-web/src/pages/ChatPage.tsx b/frontend-web/src/pages/ChatPage.tsx
index 08c09d7b..485d31cc 100644
--- a/frontend-web/src/pages/ChatPage.tsx
+++ b/frontend-web/src/pages/ChatPage.tsx
@@ -773,9 +773,9 @@ export default function ChatPage() {
                     !compactLanding &&
                     "-translate-y-[calc(50vh-96px)]",
                   compactLanding
-                    ? "max-w-(--container-width-xl)"
+                    ? "max-w-[calc(var(--container-width-xl)+150px)]"
                     : isLandingState
-                      ? "max-w-(--container-width-md)"
+                      ? "max-w-[calc(var(--container-width-md)+150px)]"
                       : "max-w-(--container-width-lg)",
                 )}
               >
diff --git a/frontend-web/src/pages/MenuManagementPage.tsx b/frontend-web/src/pages/MenuManagementPage.tsx
index 04741368..63907982 100644
--- a/frontend-web/src/pages/MenuManagementPage.tsx
+++ b/frontend-web/src/pages/MenuManagementPage.tsx
@@ -79,12 +79,10 @@ import {
   type MergedMenuNode,
 } from "@/core/page-layout/menu-overrides";
 import {
-  COMPACT_MENU_TYPES,
   COMPACT_MENU_TYPE_LABELS,
   getRuntimeCompactMenuType,
   type CompactMenuOverride,
   type CompactMenuPlacement,
-  type CompactMenuType,
 } from "@/core/page-layout/compact-menu-overrides";
 import { sidebarMenuItems, type MenuItem } from "@/core/page-layout/sidebar-menu";
 import { injectLightApps } from "@/core/page-layout/use-sidebar-menu";
@@ -510,7 +508,7 @@ export default function MenuManagementPage() {
     () => [
       {
         id: "workspace-compact-nav",
-        label: "简介模式菜单",
+        label: "简洁模式菜单",
         icon: Settings,
         children: WORKSPACE_COMPACT_NAV_ITEMS,
       },
@@ -779,7 +777,7 @@ export default function MenuManagementPage() {
           mode === "compact" ? "bg-red-600 text-white shadow-sm" : "text-slate-500 hover:text-slate-800"
         }`}
       >
-        简介模式
+        简洁模式
       </button>
     </div>
   );
@@ -980,7 +978,8 @@ export default function MenuManagementPage() {
 
 function CompactMenuManagementPanel({ baseTree }: { baseTree: MenuItem[] }) {
   const runtimeType = getRuntimeCompactMenuType();
-  const [menuType, setMenuType] = useState<CompactMenuType>(runtimeType);
+  // 不再提供 A/B 切换：固定编辑当前网页（runtime-config.js VITE_COMPACT_MENU_TYPE）使用的那套。
+  const menuType = runtimeType;
   const { data: overrides = [], isLoading } = useCompactMenuOverrides(menuType);
   const { save } = useCompactMenuOverrideActions(menuType);
   const initialRows = useMemo(() => compactRowsFromTree(baseTree, overrides), [baseTree, overrides]);
@@ -1038,23 +1037,17 @@ function CompactMenuManagementPanel({ baseTree }: { baseTree: MenuItem[] }) {
   const handleSave = async () => {
     try {
       await save.mutateAsync(currentOverrides);
-      toast.success(`简介菜单配置已保存（${COMPACT_MENU_TYPE_LABELS[menuType]}）`);
+      toast.success(`简洁菜单配置已保存（${COMPACT_MENU_TYPE_LABELS[menuType]}）`);
     } catch (err) {
       toast.error(err instanceof Error ? err.message : "保存失败");
     }
   };
 
   const handleReset = () => {
-    if (!confirm(`确定要重置${COMPACT_MENU_TYPE_LABELS[menuType]}的简介模式菜单为默认布局吗？保存后将清除该类型的自定义配置。`)) return;
+    if (!confirm(`确定要重置${COMPACT_MENU_TYPE_LABELS[menuType]}的简洁模式菜单为默认布局吗？保存后将清除该类型的自定义配置。`)) return;
     setRows(compactRowsFromTree(baseTree, []));
   };
 
-  const switchMenuType = (next: CompactMenuType) => {
-    if (next === menuType) return;
-    if (dirty && !confirm("当前类型有未保存的修改，切换后将丢失。确定切换吗？")) return;
-    setMenuType(next);
-  };
-
   return (
     <>
       <div className="relative mb-6 overflow-hidden rounded-3xl border border-slate-200/60 bg-white p-6 shadow-sm md:p-8">
@@ -1066,10 +1059,10 @@ function CompactMenuManagementPanel({ baseTree }: { baseTree: MenuItem[] }) {
             <div>
               <span className="mb-3 inline-flex select-none items-center gap-1.5 rounded-full border border-red-100 bg-red-50 px-3 py-1 text-xs font-semibold text-red-600">
                 <Sparkles className="size-3" />
-                仅简介模式生效
+                仅简洁模式生效
               </span>
               <h1 className="flex items-center gap-3 text-2xl font-bold tracking-tight text-slate-800 md:text-3xl">
-                简介菜单管理
+                简洁菜单管理
                 <span className="rounded-lg border border-slate-200/60 bg-slate-100 px-2.5 py-0.5 font-mono text-xs font-medium text-slate-500">
                   {rows.length} 个项目
                 </span>
@@ -1095,35 +1088,11 @@ function CompactMenuManagementPanel({ baseTree }: { baseTree: MenuItem[] }) {
               </button>
             </div>
           </div>
-          <div className="mb-4 flex flex-wrap items-center gap-3">
-            <div className="inline-flex rounded-2xl border border-slate-200 bg-slate-50 p-1 text-sm font-semibold">
-              {COMPACT_MENU_TYPES.map((type) => (
-                <button
-                  key={type}
-                  type="button"
-                  onClick={() => switchMenuType(type)}
-                  className={`rounded-xl px-4 py-1.5 transition-all ${
-                    menuType === type ? "bg-red-600 text-white shadow-sm" : "text-slate-500 hover:text-slate-800"
-                  }`}
-                >
-                  {COMPACT_MENU_TYPE_LABELS[type]}
-                  {runtimeType === type ? (
-                    <span className={`ml-1.5 text-[10px] font-medium ${menuType === type ? "text-white/80" : "text-slate-400"}`}>
-                      当前网页
-                    </span>
-                  ) : null}
-                </button>
-              ))}
-            </div>
-          </div>
           <p className="flex max-w-2xl items-start gap-2.5 rounded-2xl border border-slate-200/50 bg-slate-50/70 p-3.5 text-sm leading-relaxed text-slate-600">
             <Info className="mt-0.5 size-4 shrink-0 text-slate-400" />
             <span>
-              这里配置的是简介模式工作区左侧菜单，按类型 A / B 分套保存。两套网页在{" "}
-              <code className="rounded bg-slate-200/70 px-1 py-0.5 text-xs">runtime-config.js</code> 里填写{" "}
-              <code className="rounded bg-slate-200/70 px-1 py-0.5 text-xs">VITE_COMPACT_MENU_TYPE</code> 为 A 或 B
-              即可加载对应配置；全量模式菜单不受影响。当前网页正在使用
-              {COMPACT_MENU_TYPE_LABELS[runtimeType]}。
+              这里配置的是简洁模式工作区左侧菜单，保存后仅对当前网页使用的
+              {COMPACT_MENU_TYPE_LABELS[runtimeType]}生效（两套部署各改各的）；全量模式菜单不受影响。
             </span>
           </p>
         </div>
diff --git a/frontend-web/src/pages/deep-research/DeepResearchWorkbench.tsx b/frontend-web/src/pages/deep-research/DeepResearchWorkbench.tsx
index bf85a03f..ac6beb28 100644
--- a/frontend-web/src/pages/deep-research/DeepResearchWorkbench.tsx
+++ b/frontend-web/src/pages/deep-research/DeepResearchWorkbench.tsx
@@ -40,7 +40,6 @@ import {
   MessageSquareText,
   PanelRightOpen,
   RotateCcw,
-  SlidersHorizontal,
   Sparkles,
   Telescope,
   Trash2,
@@ -126,6 +125,7 @@ import {
   ReportOutlineEditor,
 } from "./ReportConfigCard";
 import { ResearchSandboxPanel } from "./ResearchSandboxPanel";
+import { WorkflowConfigButton } from "./WorkflowConfigDialog";
 import { REPORT_OUTPUT_PATH, useReportSandbox } from "./useReportSandbox";
 
 const COLLECTOR_AGENT_ID = "deep-research-collector";
@@ -1764,9 +1764,15 @@ function DeepResearchWorkbenchInner() {
                         <ImagePlus className="h-3.5 w-3.5" />配图
                       </button>
                     )}
-                    <Button variant="ghost" size="sm" className="h-7 px-2 text-xs" onClick={() => setSettingsOpen(true)} disabled={Boolean(isFollowUp) || jobWriting}>
-                      <SlidersHorizontal className="mr-1.5 h-3.5 w-3.5" />高级配置
-                    </Button>
+                    <WorkflowConfigButton
+                      variant="ghost"
+                      className="h-7 px-2 text-xs"
+                      disabled={Boolean(isFollowUp) || jobWriting}
+                      appliedContent={advancedOptions.customOutline}
+                      onApply={(content) =>
+                        setAdvancedOptions((current) => ({ ...current, customOutline: content }))
+                      }
+                    />
                   </div>
                 </div>
                 <PromptInput onSubmit={({ text }) => handleComposerSubmit(text)}>
@@ -2654,6 +2660,18 @@ function AdvancedResearchSettingsDialog({
           onChange={(value) => onChange("customOutline", value)}
         />
 
+        {/* 工作流配置：报告结构库（后台 /api/report-structures），选中卡片即填充上方报告结构 */}
+        <div className="flex items-center justify-between gap-3 rounded-xl border border-border p-3">
+          <span className="text-xs leading-5 text-muted-foreground">
+            <span className="block text-sm font-medium text-foreground">报告结构库</span>
+            在「工作流配置」中管理并选择报告结构；选中卡片后自动填充到上方报告结构。
+          </span>
+          <WorkflowConfigButton
+            appliedContent={options.customOutline}
+            onApply={(content) => onChange("customOutline", content)}
+          />
+        </div>
+
         <div className="grid gap-3 sm:grid-cols-2">
           <label className="flex cursor-pointer gap-3 rounded-xl border p-3">
             <input
diff --git a/frontend-web/src/pages/deep-research/ReportConfigCard.tsx b/frontend-web/src/pages/deep-research/ReportConfigCard.tsx
index d9065462..d7a6b721 100644
--- a/frontend-web/src/pages/deep-research/ReportConfigCard.tsx
+++ b/frontend-web/src/pages/deep-research/ReportConfigCard.tsx
@@ -32,6 +32,7 @@ import {
   type ResearchMode,
   type ResearchTone,
 } from "@/strategy-components/api/deep-research";
+import { WorkflowConfigButton } from "./WorkflowConfigDialog";
 
 const MODE_LABELS: Record<ResearchMode, string> = {
   basic: "标准",
@@ -473,6 +474,12 @@ export function ReportConfigCard({
                 onChange={(value) => update("customOutline", value)}
               />
             </div>
+            <div className="flex justify-end sm:col-span-2">
+              <WorkflowConfigButton
+                appliedContent={options.customOutline}
+                onApply={(value) => update("customOutline", value)}
+              />
+            </div>
             <label className="flex cursor-pointer items-start gap-2 rounded-lg border border-border/70 bg-background/60 p-2.5 text-[11px] leading-4 sm:col-span-2">
               <input
                 type="checkbox"
diff --git a/frontend-web/src/pages/deep-research/WorkflowConfigDialog.tsx b/frontend-web/src/pages/deep-research/WorkflowConfigDialog.tsx
new file mode 100644
index 00000000..43c4985a
--- /dev/null
+++ b/frontend-web/src/pages/deep-research/WorkflowConfigDialog.tsx
@@ -0,0 +1,549 @@
+"use client";
+
+/**
+ * 工作流配置（报告结构库）弹窗 + 触发按钮。
+ *
+ * 深度研究「高级配置」区的人口：按钮打开大弹窗，弹窗内两视图——
+ *  - 列表视图：报告结构卡片网格（后台 /api/report-structures，全局共享，
+ *    全新部署自动播种「通用分析结构 / 递进式论证结构」），支持 启用/停用 筛选
+ *    与 标题 筛选；卡片上可 switch 启停、编辑、删除（二次确认），并显示
+ *    最后修改时间与创建用户名；新增时不填状态（服务端强制默认启用）。
+ *  - 详情视图：点击卡片即「选择」该结构（内容回填到高级配置的报告结构
+ *    custom_outline），弹窗切换为展示 标题 + 接口内容，右上角「返回」回列表。
+ */
+
+import { useEffect, useMemo, useState } from "react";
+import ReactMarkdown from "react-markdown";
+import {
+  ChevronLeft,
+  Loader2,
+  Pencil,
+  Plus,
+  Search,
+  Trash2,
+  Workflow,
+  X,
+} from "lucide-react";
+import { toast } from "sonner";
+
+import { Badge } from "@/components/ui/badge";
+import { Button } from "@/components/ui/button";
+import {
+  Dialog,
+  DialogContent,
+  DialogDescription,
+  DialogFooter,
+  DialogHeader,
+  DialogTitle,
+} from "@/components/ui/dialog";
+import { Input } from "@/components/ui/input";
+import { ScrollArea } from "@/components/ui/scroll-area";
+import {
+  Select,
+  SelectContent,
+  SelectItem,
+  SelectTrigger,
+  SelectValue,
+} from "@/components/ui/select";
+import { Switch } from "@/components/ui/switch";
+import { Textarea } from "@/components/ui/textarea";
+import {
+  type ReportStructure,
+  type ReportStructureStatusFilter,
+  useReportStructures,
+  useReportStructureMutations,
+} from "@/strategy-components/api/report-structures";
+
+// ── shared helpers ───────────────────────────────────────────────────────────
+
+function formatDateTime(iso: string | null): string {
+  if (!iso) return "—";
+  const date = new Date(iso);
+  if (Number.isNaN(date.getTime())) return "—";
+  return date.toLocaleString("zh-CN", {
+    year: "numeric",
+    month: "2-digit",
+    day: "2-digit",
+    hour: "2-digit",
+    minute: "2-digit",
+  });
+}
+
+/** 创建用户名：本部署账号是邮箱，取 @ 前缀（与个人信息页「用户名」一致）。 */
+function creatorLabel(structure: ReportStructure): string {
+  if (!structure.createdBy) return "—";
+  return structure.createdBy.split("@")[0] || structure.createdBy;
+}
+
+// ── main dialog ──────────────────────────────────────────────────────────────
+
+export function WorkflowConfigDialog({
+  open,
+  onOpenChange,
+  /** 当前已应用的报告结构内容（高级配置的 custom_outline），用于高亮「当前使用」。 */
+  appliedContent,
+  /** 点击卡片选择结构时回填内容。 */
+  onApply,
+}: {
+  open: boolean;
+  onOpenChange: (open: boolean) => void;
+  appliedContent?: string;
+  onApply?: (content: string) => void;
+}) {
+  const [statusFilter, setStatusFilter] = useState<ReportStructureStatusFilter>("all");
+  const [titleFilter, setTitleFilter] = useState("");
+  const [detailId, setDetailId] = useState<string | null>(null);
+  const [editorTarget, setEditorTarget] = useState<ReportStructure | null>(null);
+  const [editorOpen, setEditorOpen] = useState(false);
+  const [deleteTarget, setDeleteTarget] = useState<ReportStructure | null>(null);
+
+  // 启用/停用筛选走服务端；标题筛选在本地做（即时、免抖动）。
+  const listQuery = useReportStructures({ status: statusFilter });
+  const mutations = useReportStructureMutations();
+
+  // 重新打开时回到列表视图。
+  useEffect(() => {
+    if (open) setDetailId(null);
+  }, [open]);
+
+  const structures = listQuery.data ?? [];
+  const visible = useMemo(() => {
+    const needle = titleFilter.trim().toLowerCase();
+    if (!needle) return structures;
+    return structures.filter((s) => s.title.toLowerCase().includes(needle));
+  }, [structures, titleFilter]);
+
+  const appliedStructureId = useMemo(() => {
+    const trimmed = (appliedContent ?? "").trim();
+    if (!trimmed) return null;
+    return structures.find((s) => s.content.trim() === trimmed)?.id ?? null;
+  }, [appliedContent, structures]);
+
+  const detail = useMemo(
+    () => structures.find((s) => s.id === detailId) ?? null,
+    [structures, detailId],
+  );
+
+  const handleSelect = (structure: ReportStructure) => {
+    onApply?.(structure.content);
+    setDetailId(structure.id);
+    toast.success(`已选择报告结构「${structure.title}」，并填充到报告结构配置`);
+  };
+
+  const handleToggleEnabled = (structure: ReportStructure, enabled: boolean) => {
+    mutations.toggleEnabled.mutate(
+      { id: structure.id, enabled },
+      {
+        onSuccess: () => toast.success(`「${structure.title}」已${enabled ? "启用" : "停用"}`),
+        onError: (err) => toast.error(err instanceof Error ? err.message : "操作失败"),
+      },
+    );
+  };
+
+  const handleDelete = async () => {
+    if (!deleteTarget) return;
+    try {
+      await mutations.remove.mutateAsync(deleteTarget.id);
+      toast.success(`已删除「${deleteTarget.title}」`);
+      if (detailId === deleteTarget.id) setDetailId(null);
+      setDeleteTarget(null);
+    } catch (err) {
+      toast.error(err instanceof Error ? err.message : "删除失败");
+    }
+  };
+
+  return (
+    <>
+      <Dialog open={open} onOpenChange={onOpenChange}>
+        <DialogContent
+          showCloseButton={false}
+          className="flex h-[85vh] max-h-[85vh] w-[calc(100vw-3rem)] max-w-5xl flex-col gap-0 overflow-hidden p-0 sm:max-w-5xl"
+        >
+          {/* header：列表视图=标题+筛选+新增；详情视图=结构标题，右上角「返回」 */}
+          {!detail ? (
+            <div className="shrink-0 border-b border-border px-5 py-4">
+              <div className="flex flex-wrap items-center gap-3">
+                <div className="min-w-0 flex-1">
+                  <h2 className="text-base font-semibold">工作流配置 · 报告结构</h2>
+                  <p className="mt-0.5 text-xs text-muted-foreground">
+                    点击卡片选择报告结构（自动填充到报告结构配置）；可启停、编辑、删除。
+                  </p>
+                </div>
+                <Button
+                  variant="ghost"
+                  size="icon"
+                  className="h-8 w-8 shrink-0"
+                  aria-label="关闭"
+                  onClick={() => onOpenChange(false)}
+                >
+                  <X className="h-4 w-4" />
+                </Button>
+              </div>
+              <div className="mt-3 flex flex-wrap items-center gap-2">
+                <div className="relative min-w-0 flex-1 sm:max-w-xs">
+                  <Search className="pointer-events-none absolute top-1/2 left-2.5 h-3.5 w-3.5 -translate-y-1/2 text-muted-foreground" />
+                  <Input
+                    value={titleFilter}
+                    onChange={(event) => setTitleFilter(event.target.value)}
+                    placeholder="按标题筛选"
+                    className="h-8 pl-8 text-xs"
+                  />
+                </div>
+                <Select
+                  value={statusFilter}
+                  onValueChange={(value) => setStatusFilter(value as ReportStructureStatusFilter)}
+                >
+                  <SelectTrigger className="h-8 w-28 shrink-0 text-xs" aria-label="启用状态筛选">
+                    <SelectValue />
+                  </SelectTrigger>
+                  <SelectContent>
+                    <SelectItem value="all">全部状态</SelectItem>
+                    <SelectItem value="enabled">已启用</SelectItem>
+                    <SelectItem value="disabled">已停用</SelectItem>
+                  </SelectContent>
+                </Select>
+                <div className="ml-auto flex shrink-0 items-center gap-2">
+                  <Badge variant="outline" className="h-6 text-[10px]">{visible.length} 套</Badge>
+                  <Button
+                    size="sm"
+                    className="h-8"
+                    onClick={() => {
+                      setEditorTarget(null);
+                      setEditorOpen(true);
+                    }}
+                  >
+                    <Plus className="mr-1 h-3.5 w-3.5" />新增结构
+                  </Button>
+                </div>
+              </div>
+            </div>
+          ) : (
+            <div className="flex shrink-0 items-center justify-between gap-3 border-b border-border px-5 py-4">
+              <div className="flex min-w-0 items-center gap-2">
+                <Badge variant="outline" className="shrink-0">报告结构</Badge>
+                <h2 className="truncate text-base font-semibold">{detail.title}</h2>
+                {detail.id === appliedStructureId && (
+                  <Badge className="shrink-0">当前使用</Badge>
+                )}
+                {!detail.enabled && <Badge variant="secondary" className="shrink-0">已停用</Badge>}
+              </div>
+              <Button variant="outline" size="sm" className="shrink-0" onClick={() => setDetailId(null)}>
+                <ChevronLeft className="mr-1 h-4 w-4" />返回
+              </Button>
+            </div>
+          )}
+
+          {/* body */}
+          {!detail ? (
+            <ScrollArea className="min-h-0 flex-1">
+              <div className="p-5">
+                {listQuery.isLoading ? (
+                  <div className="flex items-center justify-center py-16 text-sm text-muted-foreground">
+                    <Loader2 className="mr-2 h-4 w-4 animate-spin" />加载中…
+                  </div>
+                ) : listQuery.isError ? (
+                  <div className="py-16 text-center text-sm text-destructive">
+                    {(listQuery.error as Error)?.message || "加载报告结构失败"}
+                  </div>
+                ) : visible.length === 0 ? (
+                  <div className="flex flex-col items-center gap-3 py-16 text-center text-sm text-muted-foreground">
+                    <Workflow className="h-10 w-10 text-muted-foreground/30" />
+                    {titleFilter || statusFilter !== "all" ? "没有符合条件的报告结构" : "暂无报告结构，点击右上角「新增结构」创建"}
+                  </div>
+                ) : (
+                  <div className="grid gap-3 md:grid-cols-2">
+                    {visible.map((structure) => (
+                      <StructureCard
+                        key={structure.id}
+                        structure={structure}
+                        applied={structure.id === appliedStructureId}
+                        onOpen={() => handleSelect(structure)}
+                        onEdit={(event) => {
+                          event.stopPropagation();
+                          setEditorTarget(structure);
+                          setEditorOpen(true);
+                        }}
+                        onDelete={(event) => {
+                          event.stopPropagation();
+                          setDeleteTarget(structure);
+                        }}
+                        onToggleEnabled={(enabled) => handleToggleEnabled(structure, enabled)}
+                        toggling={mutations.toggleEnabled.isPending}
+                      />
+                    ))}
+                  </div>
+                )}
+              </div>
+            </ScrollArea>
+          ) : (
+            <ScrollArea className="min-h-0 flex-1">
+              <div className="mx-auto max-w-3xl p-5">
+                {detail.name && (
+                  <p className="mb-1 text-xs text-muted-foreground">描述：{detail.name}</p>
+                )}
+                <p className="mb-4 text-xs text-muted-foreground">
+                  最后修改：{formatDateTime(detail.updatedAt)} · 创建人：{creatorLabel(detail)}
+                </p>
+                <article className="prose prose-sm dark:prose-invert max-w-none rounded-xl border border-border bg-card p-5">
+                  <ReactMarkdown components={{ img: () => null }}>{detail.content}</ReactMarkdown>
+                </article>
+              </div>
+            </ScrollArea>
+          )}
+        </DialogContent>
+      </Dialog>
+
+      {/* 新增 / 编辑弹窗（嵌套） */}
+      <StructureEditorDialog
+        open={editorOpen}
+        onOpenChange={setEditorOpen}
+        target={editorTarget}
+        onSubmit={async (values) => {
+          try {
+            if (editorTarget) {
+              await mutations.update.mutateAsync({ id: editorTarget.id, patch: values });
+              toast.success("报告结构已保存");
+            } else {
+              await mutations.create.mutateAsync(values);
+              toast.success("已新增报告结构（默认启用）");
+            }
+            setEditorOpen(false);
+            setEditorTarget(null);
+          } catch (err) {
+            toast.error(err instanceof Error ? err.message : "保存失败");
+          }
+        }}
+        submitting={mutations.create.isPending || mutations.update.isPending}
+      />
+
+      {/* 删除二次确认（嵌套） */}
+      <Dialog open={deleteTarget !== null} onOpenChange={(next) => !next && setDeleteTarget(null)}>
+        <DialogContent className="max-w-sm">
+          <DialogHeader>
+            <DialogTitle>删除报告结构</DialogTitle>
+            <DialogDescription>
+              确认删除「{deleteTarget?.title}」？删除后不可恢复。
+            </DialogDescription>
+          </DialogHeader>
+          <DialogFooter>
+            <Button variant="outline" onClick={() => setDeleteTarget(null)} disabled={mutations.remove.isPending}>
+              取消
+            </Button>
+            <Button variant="destructive" onClick={() => void handleDelete()} disabled={mutations.remove.isPending}>
+              {mutations.remove.isPending && <Loader2 className="mr-1.5 h-3.5 w-3.5 animate-spin" />}
+              确认删除
+            </Button>
+          </DialogFooter>
+        </DialogContent>
+      </Dialog>
+    </>
+  );
+}
+
+// ── structure card ───────────────────────────────────────────────────────────
+
+function StructureCard({
+  structure,
+  applied,
+  onOpen,
+  onEdit,
+  onDelete,
+  onToggleEnabled,
+  toggling,
+}: {
+  structure: ReportStructure;
+  applied: boolean;
+  onOpen: () => void;
+  onEdit: (event: React.MouseEvent) => void;
+  onDelete: (event: React.MouseEvent) => void;
+  onToggleEnabled: (enabled: boolean) => void;
+  toggling: boolean;
+}) {
+  return (
+    <div
+      role="button"
+      tabIndex={0}
+      onClick={onOpen}
+      onKeyDown={(event) => {
+        if (event.key === "Enter" || event.key === " ") {
+          event.preventDefault();
+          onOpen();
+        }
+      }}
+      className={`group cursor-pointer rounded-xl border p-4 text-left transition-colors outline-none focus-visible:ring-2 focus-visible:ring-ring ${
+        applied
+          ? "border-primary/50 bg-primary/5"
+          : "border-border bg-card hover:border-primary/30 hover:bg-accent/40"
+      }`}
+    >
+      <div className="mb-1.5 flex items-start justify-between gap-2">
+        <div className="flex min-w-0 flex-wrap items-center gap-1.5">
+          <span className="truncate text-sm font-semibold">{structure.title}</span>
+          {applied && <Badge className="h-5 shrink-0 text-[10px]">当前使用</Badge>}
+          <Badge variant={structure.enabled ? "default" : "outline"} className={`h-5 shrink-0 text-[10px] ${structure.enabled ? "" : "text-muted-foreground"}`}>
+            {structure.enabled ? "启用" : "停用"}
+          </Badge>
+        </div>
+        <div className="flex shrink-0 items-center gap-1">
+          <button
+            type="button"
+            title="编辑"
+            aria-label={`编辑 ${structure.title}`}
+            className="rounded-md p-1 text-muted-foreground transition-colors hover:bg-muted hover:text-foreground"
+            onClick={onEdit}
+          >
+            <Pencil className="h-3.5 w-3.5" />
+          </button>
+          <button
+            type="button"
+            title="删除"
+            aria-label={`删除 ${structure.title}`}
+            className="rounded-md p-1 text-muted-foreground transition-colors hover:bg-destructive/10 hover:text-destructive"
+            onClick={onDelete}
+          >
+            <Trash2 className="h-3.5 w-3.5" />
+          </button>
+        </div>
+      </div>
+      {structure.name && <p className="mb-2 truncate text-xs text-muted-foreground">{structure.name}</p>}
+      <div className="flex items-center justify-between gap-3 border-t border-border/60 pt-2.5">
+        <div className="min-w-0 text-[11px] leading-4 text-muted-foreground">
+          <p className="truncate">最后修改：{formatDateTime(structure.updatedAt)}</p>
+          <p className="truncate">创建人：{creatorLabel(structure)}</p>
+        </div>
+        <div
+          className="flex shrink-0 items-center gap-1.5"
+          onClick={(event) => event.stopPropagation()}
+          role="presentation"
+        >
+          <span className="text-[11px] text-muted-foreground">{structure.enabled ? "启用" : "停用"}</span>
+          <Switch
+            checked={structure.enabled}
+            disabled={toggling}
+            aria-label={`${structure.enabled ? "停用" : "启用"} ${structure.title}`}
+            onCheckedChange={onToggleEnabled}
+          />
+        </div>
+      </div>
+    </div>
+  );
+}
+
+// ── editor dialog（新增 / 编辑共用；状态不允许填写，创建默认启用） ─────────────
+
+function StructureEditorDialog({
+  open,
+  onOpenChange,
+  target,
+  onSubmit,
+  submitting,
+}: {
+  open: boolean;
+  onOpenChange: (open: boolean) => void;
+  target: ReportStructure | null;
+  onSubmit: (values: { title: string; name: string; content: string }) => Promise<void>;
+  submitting: boolean;
+}) {
+  const [title, setTitle] = useState("");
+  const [name, setName] = useState("");
+  const [content, setContent] = useState("");
+
+  useEffect(() => {
+    if (open) {
+      setTitle(target?.title ?? "");
+      setName(target?.name ?? "");
+      setContent(target?.content ?? "");
+    }
+  }, [open, target]);
+
+  const valid = title.trim().length > 0 && content.trim().length > 0;
+
+  return (
+    <Dialog open={open} onOpenChange={onOpenChange}>
+      <DialogContent className="max-h-[85vh] w-[calc(100vw-2rem)] max-w-2xl overflow-y-auto sm:max-w-2xl">
+        <DialogHeader>
+          <DialogTitle>{target ? "编辑报告结构" : "新增报告结构"}</DialogTitle>
+          <DialogDescription>
+            {target ? "修改后将刷新「最后修改时间」。" : "新增的报告结构默认为启用状态。"}
+          </DialogDescription>
+        </DialogHeader>
+        <div className="space-y-3">
+          <label className="block">
+            <span className="mb-1.5 block text-sm font-medium">标题</span>
+            <Input
+              value={title}
+              onChange={(event) => setTitle(event.target.value)}
+              maxLength={256}
+              placeholder="例如：行业研究通用结构"
+            />
+          </label>
+          <label className="block">
+            <span className="mb-1.5 block text-sm font-medium">描述</span>
+            <Input
+              value={name}
+              onChange={(event) => setName(event.target.value)}
+              maxLength={256}
+              placeholder="可选，一句话描述这套结构的适用场景"
+            />
+          </label>
+          <label className="block">
+            <span className="mb-1.5 block text-sm font-medium">接口内容</span>
+            <Textarea
+              value={content}
+              onChange={(event) => setContent(event.target.value)}
+              maxLength={20000}
+              placeholder={"报告结构正文，支持 Markdown 标题，例如：\n# 引言\n# 一、事实与背景\n## 1.1 研究范围"}
+              className="min-h-[260px] resize-y font-mono text-xs"
+            />
+            <span className="mt-1 block text-[11px] text-muted-foreground">
+              选中卡片时，该内容会填充到高级配置的「报告结构」。
+            </span>
+          </label>
+        </div>
+        <DialogFooter>
+          <Button variant="outline" onClick={() => onOpenChange(false)} disabled={submitting}>
+            取消
+          </Button>
+          <Button disabled={!valid || submitting} onClick={() => void onSubmit({ title: title.trim(), name: name.trim(), content: content })}>
+            {submitting && <Loader2 className="mr-1.5 h-3.5 w-3.5 animate-spin" />}
+            保存
+          </Button>
+        </DialogFooter>
+      </DialogContent>
+    </Dialog>
+  );
+}
+
+// ── trigger button（自带弹窗开合状态） ────────────────────────────────────────
+
+export function WorkflowConfigButton({
+  appliedContent,
+  onApply,
+  className,
+  variant = "outline",
+  size = "sm",
+  disabled = false,
+}: {
+  appliedContent?: string;
+  onApply: (content: string) => void;
+  className?: string;
+  variant?: React.ComponentProps<typeof Button>["variant"];
+  size?: React.ComponentProps<typeof Button>["size"];
+  disabled?: boolean;
+}) {
+  const [open, setOpen] = useState(false);
+  return (
+    <>
+      <Button variant={variant} size={size} className={className} disabled={disabled} onClick={() => setOpen(true)}>
+        <Workflow className="mr-1.5 h-3.5 w-3.5" />
+        工作流配置
+      </Button>
+      <WorkflowConfigDialog
+        open={open}
+        onOpenChange={setOpen}
+        appliedContent={appliedContent}
+        onApply={onApply}
+      />
+    </>
+  );
+}
diff --git a/frontend-web/src/position-roundtable/components/PositionBusinessChainFlow.tsx b/frontend-web/src/position-roundtable/components/PositionBusinessChainFlow.tsx
index 5b65efbd..dd379170 100644
--- a/frontend-web/src/position-roundtable/components/PositionBusinessChainFlow.tsx
+++ b/frontend-web/src/position-roundtable/components/PositionBusinessChainFlow.tsx
@@ -4,7 +4,7 @@
  * 页面无关：8BF 等独立会商页可以直接复用岗位会商的纵向流程图视觉与交互，
  * 不会读取或写入岗位会商的会话、岗位节点状态。
  */
-import { useRef, useState } from "react";
+import { useEffect, useRef, useState } from "react";
 import { Check, Search } from "lucide-react";
 
 import {
@@ -13,8 +13,14 @@ import {
   DialogHeader,
   DialogTitle,
 } from "@/components/ui/dialog";
-import type { BusinessChain, ChainSeat } from "@/roundtable-planning/lib/business-chain";
+import { desensitize } from "@/lib/desensitize";
 import { cn } from "@/lib/utils";
+import {
+  positionNameById,
+  syncPositionOptionsFromRoles,
+} from "@/position-roundtable/lib/positions";
+import type { BusinessChain, ChainSeat } from "@/roundtable-planning/lib/business-chain";
+import { usePositionRoles } from "@/strategy-components/api/position-roles";
 
 const NODE_COLORS = ["#4f7df3", "#a778f5", "#65b4e8", "#f0a245", "#e675a8"];
 const FLOW_NODE_WIDTH = 184;
@@ -72,6 +78,10 @@ export function PositionBusinessChainFlow({
   submittedSeatKeys,
   onSeatClick,
 }: PositionBusinessChainFlowProps) {
+  const { data: positionRoles = [] } = usePositionRoles();
+  useEffect(() => {
+    syncPositionOptionsFromRoles(positionRoles);
+  }, [positionRoles]);
   const [dialogOpen, setDialogOpen] = useState(false);
   const submittedKeys = new Set(
     (submittedSeatKeys ?? []).map((key) => key.trim().toLowerCase()).filter(Boolean),
@@ -251,8 +261,8 @@ function BusinessChainCanvas({
                   return (
                     <FlowNode
                       key={seat.agent_id}
-                      title={seat.name || seat.agent_id}
-                      description={stage.goal?.trim() || seat.description?.trim() || "业务链席位"}
+                      title={desensitize(seat.name || seat.agent_id)}
+                      description={desensitize(positionNameById(seat.position_id))}
                       accent={accent}
                       width={nodeWidth}
                       compact={embedded}
diff --git a/frontend-web/src/position-roundtable/lib/positions.ts b/frontend-web/src/position-roundtable/lib/positions.ts
index 1ecf55dc..90c52a18 100644
--- a/frontend-web/src/position-roundtable/lib/positions.ts
+++ b/frontend-web/src/position-roundtable/lib/positions.ts
@@ -88,3 +88,31 @@ export function resolvePositionId(
 
 export const positionNameById = (positionId: string | null | undefined): string =>
   runtimePositionOptions.find((position) => position.id === positionId)?.name ?? "未分配岗位";
+
+/** Sync the flowchart/agent-config name lookup from the live 岗位角色 directory. */
+export function syncPositionOptionsFromRoles(
+  roles: ReadonlyArray<{
+    id: string;
+    name: string;
+    description: string;
+    roleType?: string;
+    enabled?: boolean;
+    isBuiltin?: boolean;
+    sortOrder?: number;
+  }>,
+): void {
+  if (!roles.length) return;
+  setRuntimePositionOptions(
+    [...roles]
+      .sort((left, right) => (left.sortOrder ?? 0) - (right.sortOrder ?? 0))
+      .map((role) => ({
+        id: role.id,
+        name: role.name,
+        description: role.description,
+        isIntentPosition: role.roleType === "intent",
+        enabled: role.enabled,
+        isBuiltin: role.isBuiltin,
+        sortOrder: role.sortOrder,
+      })),
+  );
+}
diff --git a/frontend-web/src/roundtable-planning/api/chains.ts b/frontend-web/src/roundtable-planning/api/chains.ts
index 1d8e4819..54068509 100644
--- a/frontend-web/src/roundtable-planning/api/chains.ts
+++ b/frontend-web/src/roundtable-planning/api/chains.ts
@@ -53,6 +53,8 @@ export interface ChainUpdatePayload {
   gatherFirst?: boolean;
   /** 发布开关：true 发布为公共 / false 取消发布；omit 不改。仅 owner 生效。 */
   isPublic?: boolean;
+  /** 启停开关：true 启用 / false 停用；omit 不改。仅 owner / 管理员生效。 */
+  enabled?: boolean;
 }
 
 /**
@@ -88,6 +90,7 @@ interface ServerChain {
   seat_mode?: string | null;
   gather_first?: boolean;
   is_public?: boolean;
+  enabled?: boolean;
   is_mine?: boolean;
   owner_name?: string | null;
   created_at: string | null;
@@ -106,6 +109,7 @@ function toChain(s: ServerChain): BusinessChain {
     seatMode: (s.seat_mode as SeatMode | null) ?? undefined,
     gatherFirst: s.gather_first ?? false,
     isPublic: s.is_public ?? false,
+    enabled: s.enabled ?? true,
     // 后端缺省时按「我的」处理(向后兼容旧响应)。
     isMine: s.is_mine ?? true,
     ownerName: s.owner_name ?? undefined,
@@ -141,6 +145,7 @@ function fromUpdate(p: ChainUpdatePayload) {
   if (p.seatMode !== undefined) out.seat_mode = p.seatMode;
   if (p.gatherFirst !== undefined) out.gather_first = p.gatherFirst;
   if (p.isPublic !== undefined) out.is_public = p.isPublic;
+  if (p.enabled !== undefined) out.enabled = p.enabled;
   return out;
 }
 
@@ -171,6 +176,11 @@ export async function setChainPublic(id: string, isPublic: boolean): Promise<Bus
   return updateChain(id, { isPublic });
 }
 
+/** 启用 / 停用一条链条（仅 owner / 管理员生效）。返回更新后的链条，404 返回 null。 */
+export async function setChainEnabled(id: string, enabled: boolean): Promise<BusinessChain | null> {
+  return updateChain(id, { enabled });
+}
+
 /** Fetch one chain. Returns null on 404. */
 export async function getChain(id: string): Promise<BusinessChain | null> {
   const res = await apiFetch(`${BASE()}/${encodeURIComponent(id)}`, { method: "GET" });
@@ -214,10 +224,22 @@ export async function updateChain(id: string, patch: ChainUpdatePayload): Promis
   if (res.status === 404) return null;
   const data = await readJson<ServerChain>(res, "update chain");
   const chain = toChain(data);
+  const enabledToggle = patch.enabled === true || patch.enabled === false;
   writeYLog({
-    optType: patch.isPublic === true ? OptType.enabled : patch.isPublic === false ? OptType.disabled : OptType.edit,
-    optDetail:
-      patch.isPublic === true
+    optType: enabledToggle
+      ? patch.enabled
+        ? OptType.enabled
+        : OptType.disabled
+      : patch.isPublic === true
+        ? OptType.enabled
+        : patch.isPublic === false
+          ? OptType.disabled
+          : OptType.edit,
+    optDetail: enabledToggle
+      ? patch.enabled
+        ? "CM助手-全量-业务链条-启用"
+        : "CM助手-全量-业务链条-停用"
+      : patch.isPublic === true
         ? "CM助手-全量-业务链条-发布"
         : patch.isPublic === false
           ? "CM助手-全量-业务链条-取消发布"
diff --git a/frontend-web/src/roundtable-planning/components/BusinessChainListDialog.tsx b/frontend-web/src/roundtable-planning/components/BusinessChainListDialog.tsx
index 04353b30..32824e4b 100644
--- a/frontend-web/src/roundtable-planning/components/BusinessChainListDialog.tsx
+++ b/frontend-web/src/roundtable-planning/components/BusinessChainListDialog.tsx
@@ -9,9 +9,10 @@
  * 卡片席位名优先用实时 agent 数据（`listAgents()`，与 `BusinessChainPicker` 一致），
  * agent 改名后此弹窗也能同步显示；解析不到才回退到落库快照（`seat.name`）。
  */
-import { Fragment, useCallback, useEffect, useState } from "react";
+import { Fragment, useCallback, useEffect, useMemo, useState } from "react";
 import {
   ArrowRightIcon,
+  ClockIcon,
   GlobeIcon,
   LayersIcon,
   Loader2Icon,
@@ -28,14 +29,15 @@ import {
   DialogHeader,
   DialogTitle,
 } from "@/components/ui/dialog";
+import { Switch } from "@/components/ui/switch";
 import { cn } from "@/lib/utils";
 import { useDesensitizeSubtree } from "@/roundtable-planning/lib/use-desensitize-subtree";
 
 import { getStoredAuth } from "@/core/auth";
 import { listAgents } from "@/core/agents";
 
-import { listChains, setChainPublic, type ChainScope } from "../api/chains";
-import { chainTagClasses, type BusinessChain } from "../lib/business-chain";
+import { listChains, setChainEnabled, setChainPublic, type ChainScope } from "../api/chains";
+import { chainTagClasses, isChainEnabled, type BusinessChain } from "../lib/business-chain";
 import { getAvatarTypeFor } from "../lib/roundtable-constants";
 
 export interface BusinessChainListDialogProps {
@@ -57,6 +59,8 @@ export interface BusinessChainListBodyProps {
   emptyHint?: string;
 }
 
+export type EnabledFilter = "all" | "enabled" | "disabled";
+
 /**
  * 业务链条列表主体（筛选 tab + 滚动卡片列表）。
  *
@@ -76,7 +80,10 @@ export function BusinessChainListBody({
   const isAdmin = getStoredAuth()?.system_role === "admin";
   // 筛选：我的 / 公共（已发布）/（管理员）全部。
   const [scope, setScope] = useState<ChainScope>("mine");
+  // 启停筛选：全部 / 启用 / 停用。默认全部，关掉开关后卡片仍留在当前列表。
+  const [enabledFilter, setEnabledFilter] = useState<EnabledFilter>("all");
   const [publishingId, setPublishingId] = useState<string | null>(null);
+  const [togglingId, setTogglingId] = useState<string | null>(null);
   // 实时名称表（id 小写 → 最新 name）：用于把席位的落库快照名刷新成最新（agent 改名同步）。
   const [nameById, setNameById] = useState<Map<string, string>>(new Map());
 
@@ -137,8 +144,12 @@ export function BusinessChainListBody({
     async (chain: BusinessChain) => {
       setPublishingId(chain.id);
       try {
-        await setChainPublic(chain.id, !chain.isPublic);
-        await refresh(scope);
+        const updated = await setChainPublic(chain.id, !chain.isPublic);
+        if (updated) {
+          setChains((prev) => prev.map((c) => (c.id === updated.id ? updated : c)));
+        } else {
+          await refresh(scope);
+        }
       } catch (err) {
         setError(err instanceof Error ? err.message : String(err));
       } finally {
@@ -148,25 +159,69 @@ export function BusinessChainListBody({
     [refresh, scope],
   );
 
+  const handleToggleEnabled = useCallback(
+    async (chain: BusinessChain) => {
+      setTogglingId(chain.id);
+      try {
+        const updated = await setChainEnabled(chain.id, !isChainEnabled(chain));
+        if (updated) {
+          setChains((prev) => prev.map((c) => (c.id === updated.id ? updated : c)));
+        } else {
+          await refresh(scope);
+        }
+      } catch (err) {
+        setError(err instanceof Error ? err.message : String(err));
+      } finally {
+        setTogglingId(null);
+      }
+    },
+    [refresh, scope],
+  );
+
+  const visibleChains = useMemo(() => {
+    if (enabledFilter === "enabled") return chains.filter((c) => isChainEnabled(c));
+    if (enabledFilter === "disabled") return chains.filter((c) => !isChainEnabled(c));
+    return chains;
+  }, [chains, enabledFilter]);
+
   return (
     <>
-      {/* 筛选：我的 / 公共 /（管理员）全部 */}
-      <div className="shrink-0 flex items-center gap-1 rounded-lg border border-border bg-muted/40 p-0.5 self-start">
-        {((isAdmin ? ["mine", "public", "all"] : ["mine", "public"]) as ChainScope[]).map((s) => (
-          <button
-            key={s}
-            type="button"
-            onClick={() => setScope(s)}
-            className={cn(
-              "rounded-md px-3 py-1 text-xs font-medium transition",
-              scope === s
-                ? "bg-primary text-primary-foreground shadow-sm"
-                : "text-muted-foreground hover:text-foreground",
-            )}
-          >
-            {s === "mine" ? "我的" : s === "public" ? "公共" : "全部"}
-          </button>
-        ))}
+      {/* 筛选：我的 / 公共 /（管理员）全部  +  全部 / 启用 / 停用 */}
+      <div className="flex shrink-0 flex-wrap items-center gap-2">
+        <div className="flex items-center gap-1 rounded-lg border border-border bg-muted/40 p-0.5 self-start">
+          {((isAdmin ? ["mine", "public", "all"] : ["mine", "public"]) as ChainScope[]).map((s) => (
+            <button
+              key={s}
+              type="button"
+              onClick={() => setScope(s)}
+              className={cn(
+                "rounded-md px-3 py-1 text-xs font-medium transition",
+                scope === s
+                  ? "bg-primary text-primary-foreground shadow-sm"
+                  : "text-muted-foreground hover:text-foreground",
+              )}
+            >
+              {s === "mine" ? "我的" : s === "public" ? "公共" : "全部"}
+            </button>
+          ))}
+        </div>
+        <div className="flex items-center gap-1 rounded-lg border border-border bg-muted/40 p-0.5 self-start">
+          {(["all", "enabled", "disabled"] as EnabledFilter[]).map((f) => (
+            <button
+              key={f}
+              type="button"
+              onClick={() => setEnabledFilter(f)}
+              className={cn(
+                "rounded-md px-3 py-1 text-xs font-medium transition",
+                enabledFilter === f
+                  ? "bg-primary text-primary-foreground shadow-sm"
+                  : "text-muted-foreground hover:text-foreground",
+              )}
+            >
+              {f === "all" ? "全部状态" : f === "enabled" ? "启用" : "停用"}
+            </button>
+          ))}
+        </div>
       </div>
 
       <div className="custom-scrollbar -mr-1 min-h-0 flex-1 overflow-y-auto py-3 pr-1">
@@ -179,30 +234,40 @@ export function BusinessChainListBody({
           <div className="rounded-md border border-destructive/30 bg-destructive/5 p-4 text-sm text-destructive">
             加载业务链条失败：{error}
           </div>
-        ) : chains.length === 0 ? (
+        ) : visibleChains.length === 0 ? (
           <div className="flex flex-col items-center justify-center gap-3 px-6 py-12 text-center">
             <LayersIcon className="size-10 text-primary/40" />
             <div className="text-sm font-semibold">
-              {scope === "public" ? "还没有公共链条" : "还没有业务链条"}
+              {enabledFilter !== "all"
+                ? enabledFilter === "enabled"
+                  ? "没有启用的业务链条"
+                  : "没有停用的业务链条"
+                : scope === "public"
+                  ? "还没有公共链条"
+                  : "还没有业务链条"}
             </div>
             <p className="max-w-sm text-xs leading-relaxed text-muted-foreground">
-              {scope === "public"
-                ? "公共链条是其他账号（或你自己）发布出来、所有人可见的链条。去「我的」里发布一条试试。"
-                : emptyHint ??
-                  "业务链条是可复用的有序智能体编排。点击下方「配置」去创建第一条链条。"}
+              {enabledFilter !== "all"
+                ? "试试切换「全部状态」，或到卡片上用开关切换启停。"
+                : scope === "public"
+                  ? "公共链条是其他账号（或你自己）发布出来、所有人可见的链条。去「我的」里发布一条试试。"
+                  : emptyHint ??
+                    "业务链条是可复用的有序智能体编排。点击下方「配置」去创建第一条链条。"}
             </p>
           </div>
         ) : (
           <div className="space-y-3.5">
-            {chains.map((chain) => (
+            {visibleChains.map((chain) => (
               <ChainRow
                 key={chain.id}
                 chain={chain}
                 canManage={isAdmin || chain.isMine !== false}
                 nameById={nameById}
                 publishing={publishingId === chain.id}
+                toggling={togglingId === chain.id}
                 onSettings={() => onConfigure(chain.id)}
                 onTogglePublic={() => handleTogglePublic(chain)}
+                onToggleEnabled={() => handleToggleEnabled(chain)}
               />
             ))}
           </div>
@@ -252,22 +317,33 @@ function ChainRow({
   canManage,
   nameById,
   publishing,
+  toggling,
   onSettings,
   onTogglePublic,
+  onToggleEnabled,
 }: {
   chain: BusinessChain;
-  /** 能否编辑 / 发布（本人链条 或 管理员）。 */
+  /** 能否编辑 / 发布 / 启停（本人链条 或 管理员）。 */
   canManage: boolean;
   nameById: Map<string, string>;
   publishing: boolean;
+  toggling: boolean;
   onSettings: () => void;
   onTogglePublic: () => void;
+  onToggleEnabled: () => void;
 }) {
   // 配置者用户名：后端返回 email，展示时取 @ 前的本地部分更像「用户名」。
   const ownerName = chain.ownerName ? chain.ownerName.split("@")[0] : "";
+  const enabled = isChainEnabled(chain);
+  const updatedLabel = formatUpdatedAt(chain.updatedAt);
   return (
-    <div className="rounded-2xl border border-border bg-card p-5 transition-colors hover:border-primary/40">
-      {/* 标题行 + 公共徽标 + 发布/设置按钮 */}
+    <div
+      className={cn(
+        "rounded-2xl border border-border bg-card p-5 transition-colors hover:border-primary/40",
+        !enabled && "opacity-70",
+      )}
+    >
+      {/* 标题行 + 公共/停用徽标 + 启停开关 / 发布 / 设置按钮 */}
       <div className="flex items-center gap-2.5">
         <div className="truncate text-base font-semibold">{chain.title || "未命名链条"}</div>
         <span className="shrink-0 rounded-full bg-muted px-2 py-0.5 text-xs font-medium tabular-nums text-muted-foreground">
@@ -279,7 +355,26 @@ function ChainRow({
             公共
           </span>
         )}
+        {!enabled && (
+          <span className="shrink-0 rounded-full bg-muted px-2 py-0.5 text-xs font-medium text-muted-foreground">
+            已停用
+          </span>
+        )}
         <div className="ml-auto flex shrink-0 items-center gap-1.5">
+          {/* 启停：本人的链条或管理员可操作。 */}
+          {canManage && (
+            <label className="flex items-center gap-1.5 pr-1 text-xs text-muted-foreground">
+              <Switch
+                checked={enabled}
+                disabled={toggling}
+                onCheckedChange={() => onToggleEnabled()}
+                aria-label={enabled ? "停用该业务链条" : "启用该业务链条"}
+              />
+              <span className={cn("min-w-[2em]", enabled ? "text-foreground" : "text-muted-foreground")}>
+                {enabled ? "启用" : "停用"}
+              </span>
+            </label>
+          )}
           {/* 发布 / 取消发布：本人的链条或管理员可操作。 */}
           {canManage && (
             <Button
@@ -312,13 +407,20 @@ function ChainRow({
         </div>
       </div>
 
-      {ownerName && (
-        <div
-          className="mt-1.5 flex items-center gap-1 text-xs text-muted-foreground"
-          title={chain.ownerName}
-        >
-          <UserIcon className="size-3.5 shrink-0" />
-          <span className="truncate">配置人：{ownerName}</span>
+      {(ownerName || updatedLabel) && (
+        <div className="mt-1.5 flex flex-wrap items-center gap-x-3 gap-y-1 text-xs text-muted-foreground">
+          {ownerName && (
+            <div className="flex items-center gap-1" title={chain.ownerName}>
+              <UserIcon className="size-3.5 shrink-0" />
+              <span className="truncate">配置人：{ownerName}</span>
+            </div>
+          )}
+          {updatedLabel && (
+            <div className="flex items-center gap-1" title={chain.updatedAt}>
+              <ClockIcon className="size-3.5 shrink-0" />
+              <span>最后修改：{updatedLabel}</span>
+            </div>
+          )}
         </div>
       )}
 
@@ -401,3 +503,21 @@ function ChainSeatsFlow({
     </div>
   );
 }
+
+/** 后端 `updated_at` 形如 `2026-08-19T17:50:00+08:00`；展示到分钟。 */
+function formatUpdatedAt(iso: string | undefined): string {
+  if (!iso) return "";
+  const date = new Date(iso);
+  if (!Number.isNaN(date.getTime())) {
+    return date.toLocaleString("zh-CN", {
+      year: "numeric",
+      month: "2-digit",
+      day: "2-digit",
+      hour: "2-digit",
+      minute: "2-digit",
+      hour12: false,
+    });
+  }
+  const s = iso.replace("T", " ");
+  return s.length >= 16 ? s.slice(0, 16) : s;
+}
diff --git a/frontend-web/src/roundtable-planning/components/BusinessChainPicker.tsx b/frontend-web/src/roundtable-planning/components/BusinessChainPicker.tsx
index 11498d84..af4baaff 100644
--- a/frontend-web/src/roundtable-planning/components/BusinessChainPicker.tsx
+++ b/frontend-web/src/roundtable-planning/components/BusinessChainPicker.tsx
@@ -34,6 +34,7 @@ import { listChains, type ChainScope } from "../api/chains";
 import {
   CHAIN_MIN_SEATS,
   chainTagClasses,
+  isChainEnabled,
   type BusinessChain,
   type SeatSelectionResult,
 } from "../lib/business-chain";
@@ -77,7 +78,7 @@ export function BusinessChainPicker({
           listAgents().catch(() => null), // agent 列表失败不阻断链条展示
         ]);
         if (cancelled) return;
-        setChains(chainList);
+        setChains(chainList.filter(isChainEnabled));
         if (agents) {
           const candidates = filterRecommendCandidates(agents);
           setExistingIds(new Set(candidates.map((c) => c.agent_id)));
diff --git a/frontend-web/src/roundtable-planning/lib/business-chain.ts b/frontend-web/src/roundtable-planning/lib/business-chain.ts
index 18d4e953..3f29f2d0 100644
--- a/frontend-web/src/roundtable-planning/lib/business-chain.ts
+++ b/frontend-web/src/roundtable-planning/lib/business-chain.ts
@@ -95,6 +95,11 @@ export interface BusinessChain {
   gatherFirst?: boolean;
   /** 是否已发布为公共（所有账号可在「公共」筛选查看 / 选入研讨）。 */
   isPublic?: boolean;
+  /**
+   * 启停状态。缺省 / undefined = 启用（存量链条无该字段时按启用处理）。
+   * 停用后管理弹窗仍可见，选链弹窗 / 公共入口不再选用。
+   */
+  enabled?: boolean;
   /** 是否当前用户所有（仅本人可编辑 / 删除 / 发布；公共筛选里别人的链条为 false）。 */
   isMine?: boolean;
   /** 配置者（owner）的可读用户名（后端解析的 email）；解析失败时为空。 */
@@ -151,6 +156,11 @@ export interface SeatSelectionResult {
 export const CHAIN_MIN_SEATS = 2;
 export const CHAIN_MAX_SEATS = 20;
 
+/** 启停状态：缺省 / undefined 视为启用（存量链条无该字段）。 */
+export function isChainEnabled(chain: Pick<BusinessChain, "enabled">): boolean {
+  return chain.enabled !== false;
+}
+
 /**
  * 由一条业务链条 + 当前候选智能体，构造 `SeatSelectionResult`。
  *
diff --git a/frontend-web/src/roundtable-planning/pages/RoundtablePlanningPage.tsx b/frontend-web/src/roundtable-planning/pages/RoundtablePlanningPage.tsx
index e3ea2bfe..831da692 100644
--- a/frontend-web/src/roundtable-planning/pages/RoundtablePlanningPage.tsx
+++ b/frontend-web/src/roundtable-planning/pages/RoundtablePlanningPage.tsx
@@ -134,6 +134,7 @@ import {
 } from "@/roundtable-planning/lib/seat-filter";
 import {
   buildChainSeatSelection,
+  isChainEnabled,
   type BusinessChain,
   type ChainSeat,
   type SeatSelectionResult,
@@ -3395,7 +3396,8 @@ function RoundtableEntryRail({
     void listChains("public")
       .then((all) => {
         if (!cancelled) {
-          setChains(lockBusinessChain ? all.filter((chain) => chain.id === lockedChainId) : all);
+          const usable = all.filter(isChainEnabled);
+          setChains(lockBusinessChain ? usable.filter((chain) => chain.id === lockedChainId) : usable);
         }
       })
       .catch(() => {
diff --git a/frontend-web/src/strategy-components/api/report-structures.ts b/frontend-web/src/strategy-components/api/report-structures.ts
new file mode 100644
index 00000000..4a129f30
--- /dev/null
+++ b/frontend-web/src/strategy-components/api/report-structures.ts
@@ -0,0 +1,180 @@
+/**
+ * 深度研究「工作流配置 / 报告结构」API 客户端 + React Query hooks。
+ *
+ * 后端：`/api/report-structures`（见 app/gateway/routers/report_structures.py）。
+ * 全局共享、无用户隔离；新增永远默认启用（服务端强制，不收 enabled 字段）。
+ * snake_case ⇄ camelCase 在此边界收敛。
+ */
+
+import {
+  keepPreviousData,
+  useMutation,
+  useQuery,
+  useQueryClient,
+} from "@tanstack/react-query";
+
+import { apiFetch } from "@/core/api/fetch-client";
+import { getBackendBaseURL } from "@/core/config";
+
+export interface ReportStructure {
+  id: string;
+  /** 标题：卡片主标签，也是列表「标题筛选」的目标。 */
+  title: string;
+  /** 描述：一句话说明这套结构的适用场景（存储列沿用 name）。 */
+  name: string;
+  /** 接口内容：选中卡片后应用到报告结构（custom_outline）的 Markdown 文本。 */
+  content: string;
+  enabled: boolean;
+  /** 创建用户名（本部署为邮箱；系统内置种子为“系统内置”）。 */
+  createdBy: string | null;
+  createdAt: string | null;
+  /** 最后修改时间（编辑、启停都会刷新）。 */
+  updatedAt: string | null;
+}
+
+export type ReportStructureInput = {
+  title: string;
+  name: string;
+  content: string;
+};
+
+export type ReportStructurePatch = Partial<ReportStructureInput>;
+
+export type ReportStructureStatusFilter = "all" | "enabled" | "disabled";
+
+const BASE = () => `${getBackendBaseURL()}/api/report-structures`;
+
+interface ServerStructure {
+  id: string;
+  title: string;
+  name: string;
+  content: string;
+  enabled: boolean;
+  created_by: string | null;
+  created_at: string | null;
+  updated_at: string | null;
+}
+
+function toStructure(s: ServerStructure): ReportStructure {
+  return {
+    id: s.id,
+    title: s.title,
+    name: s.name ?? "",
+    content: s.content,
+    enabled: s.enabled ?? true,
+    createdBy: s.created_by ?? null,
+    createdAt: s.created_at ?? null,
+    updatedAt: s.updated_at ?? null,
+  };
+}
+
+async function readJson<T>(res: Response, ctx: string): Promise<T> {
+  if (!res.ok) {
+    const data = await res.json().catch(() => null as unknown);
+    const detail =
+      data && typeof data === "object" && "detail" in data
+        ? String((data as { detail: unknown }).detail)
+        : `${ctx} failed (${res.status})`;
+    throw new Error(detail);
+  }
+  return (await res.json()) as T;
+}
+
+export async function listReportStructures(params: {
+  status?: ReportStructureStatusFilter;
+  title?: string;
+} = {}): Promise<ReportStructure[]> {
+  const qs = new URLSearchParams();
+  if (params.status === "enabled") qs.set("enabled", "true");
+  if (params.status === "disabled") qs.set("enabled", "false");
+  const title = params.title?.trim();
+  if (title) qs.set("title", title);
+  const query = qs.toString();
+  const res = await apiFetch(`${BASE()}${query ? `?${query}` : ""}`, { method: "GET" });
+  const data = await readJson<{ structures: ServerStructure[] }>(res, "list report structures");
+  return (data.structures ?? []).map(toStructure);
+}
+
+export async function createReportStructure(input: ReportStructureInput): Promise<ReportStructure> {
+  const res = await apiFetch(BASE(), {
+    method: "POST",
+    headers: { "Content-Type": "application/json" },
+    body: JSON.stringify({ title: input.title, name: input.name, content: input.content }),
+  });
+  return toStructure(await readJson<ServerStructure>(res, "create report structure"));
+}
+
+export async function updateReportStructure(
+  id: string,
+  patch: ReportStructurePatch,
+): Promise<ReportStructure> {
+  const body: Record<string, unknown> = {};
+  if (patch.title !== undefined) body.title = patch.title;
+  if (patch.name !== undefined) body.name = patch.name;
+  if (patch.content !== undefined) body.content = patch.content;
+  const res = await apiFetch(`${BASE()}/${encodeURIComponent(id)}`, {
+    method: "PUT",
+    headers: { "Content-Type": "application/json" },
+    body: JSON.stringify(body),
+  });
+  return toStructure(await readJson<ServerStructure>(res, "update report structure"));
+}
+
+export async function setReportStructureEnabled(id: string, enabled: boolean): Promise<ReportStructure> {
+  const res = await apiFetch(`${BASE()}/${encodeURIComponent(id)}/enabled`, {
+    method: "PUT",
+    headers: { "Content-Type": "application/json" },
+    body: JSON.stringify({ enabled }),
+  });
+  return toStructure(await readJson<ServerStructure>(res, "toggle report structure"));
+}
+
+export async function deleteReportStructure(id: string): Promise<void> {
+  const res = await apiFetch(`${BASE()}/${encodeURIComponent(id)}`, { method: "DELETE" });
+  // 204 无响应体，不能再走 readJson 的 res.json()。
+  if (!res.ok) {
+    const data = (await res.json().catch(() => null)) as { detail?: unknown } | null;
+    throw new Error(
+      data && typeof data.detail === "string" ? data.detail : `delete report structure failed (${res.status})`,
+    );
+  }
+}
+
+// ── React Query hooks（工作流配置弹窗内各视图共用缓存） ──────────────────────
+
+const REPORT_STRUCTURES_KEY = "report-structures";
+
+export function useReportStructures(params: { status?: ReportStructureStatusFilter; title?: string } = {}) {
+  return useQuery({
+    queryKey: [REPORT_STRUCTURES_KEY, "list", params.status ?? "all", params.title ?? ""],
+    queryFn: () => listReportStructures(params),
+    placeholderData: keepPreviousData,
+    refetchOnWindowFocus: false,
+  });
+}
+
+export function useReportStructureMutations() {
+  const queryClient = useQueryClient();
+  const invalidate = () => queryClient.invalidateQueries({ queryKey: [REPORT_STRUCTURES_KEY] });
+
+  const create = useMutation({
+    mutationFn: (input: ReportStructureInput) => createReportStructure(input),
+    onSuccess: invalidate,
+  });
+  const update = useMutation({
+    mutationFn: ({ id, patch }: { id: string; patch: ReportStructurePatch }) =>
+      updateReportStructure(id, patch),
+    onSuccess: invalidate,
+  });
+  const toggleEnabled = useMutation({
+    mutationFn: ({ id, enabled }: { id: string; enabled: boolean }) =>
+      setReportStructureEnabled(id, enabled),
+    onSuccess: invalidate,
+  });
+  const remove = useMutation({
+    mutationFn: (id: string) => deleteReportStructure(id),
+    onSuccess: invalidate,
+  });
+
+  return { create, update, toggleEnabled, remove };
+}
diff --git a/offline-backend-20260512/backend/CLAUDE.md b/offline-backend-20260512/backend/CLAUDE.md
index 58883967..eb21ba3b 100644
--- a/offline-backend-20260512/backend/CLAUDE.md
+++ b/offline-backend-20260512/backend/CLAUDE.md
@@ -482,6 +482,16 @@ Backs the zzJC / zzXD agent-chat left sidebar (「按类型聚合智能体 + 聊
 
 ### Position-collaboration foundation
 
+### Roundtable business-chain enable/disable
+
+`roundtable_chains.enabled` (BOOLEAN NOT NULL DEFAULT 1, migration `20260819_02`)
+is the per-chain 启停 switch. New and existing chains default to **enabled**.
+Owner or admin can flip it via `PUT /api/roundtable-chains/{id}` `{enabled}`.
+Disabled chains stay visible in the management list (with an 启用/停用 filter)
+and still show `updated_at`; the frontend hides them from the picker / public
+entry rail so they cannot be selected into a roundtable. Tests:
+`tests/test_roundtable_chain_stages.py` (`test_create_defaults_enabled_then_toggle`).
+
 ### Roundtable business-chain human validation
 
 The existing `/api/roundtable-chains` `seats` JSON also accepts the optional
diff --git a/offline-backend-20260512/backend/app/gateway/app.py b/offline-backend-20260512/backend/app/gateway/app.py
index b1316762..1bb4e1b5 100644
--- a/offline-backend-20260512/backend/app/gateway/app.py
+++ b/offline-backend-20260512/backend/app/gateway/app.py
@@ -1,1526 +1,1527 @@
-import asyncio
-import logging
-import os
-import queue
-import sys
-import threading
-from collections.abc import AsyncGenerator
-from contextlib import asynccontextmanager
-from datetime import datetime
-from time import perf_counter
-
-from dotenv import load_dotenv
-from fastapi import FastAPI
-from fastapi.middleware.cors import CORSMiddleware
-
-# Load .env BEFORE any module that reads os.environ at import time.
-# Without this, GATEWAY_CORS_ORIGINS / AUTH_JWT_SECRET / DEER_FLOW_AUTH_DISABLED
-# placed in /workspace/.env are silently ignored when the container is started
-# without explicit `-e` flags, leading to ephemeral JWT secrets, missing CORS
-# origins, and "Failed to fetch" in the browser.
-load_dotenv()
-
-from app.gateway.auth.disabled_mode import is_auth_disabled  # noqa: E402
-from app.gateway.auth_middleware import AuthMiddleware
-from app.gateway.config import get_gateway_config
-from app.gateway.csrf_middleware import CSRFMiddleware
-from app.gateway.deps import langgraph_runtime
-from app.gateway.routers import (
-    ai_writing,
-    admin_active_runs,
-    admin_users,
-    agents,
-    artifacts,
-    assistants_compat,
-    auth,
-    browser_context,
-    business_mapping,
+import asyncio
+import logging
+import os
+import queue
+import sys
+import threading
+from collections.abc import AsyncGenerator
+from contextlib import asynccontextmanager
+from datetime import datetime
+from time import perf_counter
+
+from dotenv import load_dotenv
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+# Load .env BEFORE any module that reads os.environ at import time.
+# Without this, GATEWAY_CORS_ORIGINS / AUTH_JWT_SECRET / DEER_FLOW_AUTH_DISABLED
+# placed in /workspace/.env are silently ignored when the container is started
+# without explicit `-e` flags, leading to ephemeral JWT secrets, missing CORS
+# origins, and "Failed to fetch" in the browser.
+load_dotenv()
+
+from app.gateway.auth.disabled_mode import is_auth_disabled  # noqa: E402
+from app.gateway.auth_middleware import AuthMiddleware
+from app.gateway.config import get_gateway_config
+from app.gateway.csrf_middleware import CSRFMiddleware
+from app.gateway.deps import langgraph_runtime
+from app.gateway.routers import (
+    ai_writing,
+    admin_active_runs,
+    admin_users,
+    agents,
+    artifacts,
+    assistants_compat,
+    auth,
+    browser_context,
+    business_mapping,
     channels,
     checkpoint_migration,
     compact_menu_overrides,
     feedback,
-    dashboard_sessions,
-    deep_research,
-    html_page_favorites,
-    intent,
-    knowledge,
-    knowledge_ingest,
-    light_apps,
-    llm_metrics,
-    llmwiki,
-    mcp,
-    memory,
-    menu_overrides,
-    model_management,
-    models,
-    multi_agent,
-    notifications,
-    open_3qfx,
-    open_chat,
-    parallel_agent_tasks,
-    parallel_agents,
-    position_roundtable,
-    positions,
-    position_roles,
-    public_config,
-    public_embed,
-    public_leaderboard,
-    public_scheduled_tasks,
-    recommend,
-    flow_extract,
-    recommended_questions,
-    roundtable_chains,
-    roundtable_artifact_submissions,
-    roundtable_diagnostics,
-    roundtable_draft_shares,
-    roundtable_drafts,
-    roundtable_jobs,
-    roundtable_task_drafts,
-    runs,
-    scheduled_tasks,
-    sensitive_words,
-    skill_addresses,
-    skills,
-    suggestions,
-    system_settings,
-    tags,
-    task_buttons,
-    taskcop_tasks,
-    task_reports,
-    thread_runs,
-    tool_metrics,
-    thread_shares,
-    threads,
-    uploads,
-    user_preferences,
-    user_prompts,
-    writing,
-)
-from deerflow.config import app_config as deerflow_app_config
-from deerflow.config.app_config import apply_logging_level
-from deerflow.runtime.scheduler import ScheduledTaskService, set_scheduled_task_service
-
-AppConfig = deerflow_app_config.AppConfig
-get_app_config = deerflow_app_config.get_app_config
-
-# Default logging; lifespan overrides from config.yaml log_level.
-logging.basicConfig(
-    level=logging.INFO,
-    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
-    datefmt="%Y-%m-%d %H:%M:%S",
-)
-
-logger = logging.getLogger(__name__)
-
-# Upper bound (seconds) each lifespan shutdown hook is allowed to run.
-# Bounds worker exit time so uvicorn's reload supervisor does not keep
-# firing signals into a worker that is stuck waiting for shutdown cleanup.
-_SHUTDOWN_HOOK_TIMEOUT_SECONDS = 5.0
-
-
-async def _ensure_admin_user(app: FastAPI) -> None:
-    """Startup hook: handle first boot and migrate orphan threads otherwise.
-
-    After admin creation, migrate orphan threads from the LangGraph
-    store (metadata.user_id unset) to the admin account. This is the
-    "no-auth ? with-auth" upgrade path: users who ran DeerFlow without
-    authentication have existing LangGraph thread data that needs an
-    owner assigned.
-        First boot (no admin exists):
-            - Does NOT create any user accounts automatically.
-            - The operator must visit ``/setup`` to create the first admin.
-
-    Subsequent boots (admin already exists):
-      - Runs the one-time "no-auth ? with-auth" orphan thread migration for
-        existing LangGraph thread metadata that has no owner_id.
-
-    No SQL persistence migration is needed: the four user_id columns
-    (threads_meta, runs, run_events, feedback) only come into existence
-    alongside the auth module via create_all, so freshly created tables
-    never contain NULL-owner rows.
-    """
-    from sqlalchemy import select
-
-    from app.gateway.deps import get_local_provider
-    from deerflow.persistence.engine import get_session_factory
-    from deerflow.persistence.user.model import UserRow
-
-    try:
-        provider = get_local_provider()
-    except RuntimeError:
-        # Auth persistence may not be initialized in some test/boot paths.
-        # Skip admin migration work rather than failing gateway startup.
-        logger.warning("Auth persistence not ready; skipping admin bootstrap check")
-        return
-
-    sf = get_session_factory()
-    if sf is None:
-        return
-
-    admin_count = await provider.count_admin_users()
-
-    if admin_count == 0:
-        logger.info("=" * 60)
-        logger.info("  First boot detected ? no admin account exists.")
-        logger.info("  Visit /setup to complete admin account creation.")
-        logger.info("=" * 60)
-        return
-
-    # Admin already exists ? run orphan thread migration for any
-    # LangGraph thread metadata that pre-dates the auth module.
-    async with sf() as session:
-        stmt = select(UserRow).where(UserRow.system_role == "admin").limit(1)
-        row = (await session.execute(stmt)).scalar_one_or_none()
-
-    if row is None:
-        return  # Should not happen (admin_count > 0 above), but be safe.
-
-    admin_id = str(row.id)
-
-    # LangGraph store orphan migration ? non-fatal.
-    # This covers the "no-auth ? with-auth" upgrade path for users
-    # whose existing LangGraph thread metadata has no user_id set.
-    store = getattr(app.state, "store", None)
-    if store is not None:
-        try:
-            migrated = await _migrate_orphaned_threads(store, admin_id)
-            if migrated:
-                logger.info("Migrated %d orphan LangGraph thread(s) to admin", migrated)
-        except Exception:
-            logger.exception("LangGraph thread migration failed (non-fatal)")
-
-
-async def _iter_store_items(store, namespace, *, page_size: int = 500):
-    """Paginated async iterator over a LangGraph store namespace.
-
-    Replaces the old hardcoded ``limit=1000`` call with a cursor-style
-    loop so that environments with more than one page of orphans do
-    not silently lose data. Terminates when a page is empty OR when a
-    short page arrives (indicating the last page).
-    """
-    offset = 0
-    while True:
-        batch = await store.asearch(namespace, limit=page_size, offset=offset)
-        if not batch:
-            return
-        for item in batch:
-            yield item
-        if len(batch) < page_size:
-            return
-        offset += page_size
-
-
-async def _sync_legacy_skills(app: FastAPI) -> None:
-    """Idempotently register existing on-disk ``skills/custom/*`` directories
-    in the ``skills`` DB table, owned by the first admin and published=True.
-
-    Without this step, after upgrading to per-user skill ownership the existing
-    custom skill folders would not be visible to anyone (the SkillStore
-    visibility filter relies on a DB row). Running this once on every boot is
-    cheap (folder scan + one ``ensure_legacy`` per name) and idempotent.
-
-    Non-fatal: a failure here MUST NOT block server startup.
-    """
-    from sqlalchemy import select
-
-    from deerflow.persistence.engine import get_session_factory
-    from deerflow.persistence.user.model import UserRow
-    from deerflow.skills.storage import get_or_new_skill_storage
-    from deerflow.skills.types import SkillCategory
-
-    sf = get_session_factory()
-    if sf is None:
-        return
-
-    skill_store = getattr(app.state, "skill_store", None)
-    if skill_store is None:
-        return
-
-    async with sf() as session:
-        stmt = select(UserRow).where(UserRow.system_role == "admin").limit(1)
-        row = (await session.execute(stmt)).scalar_one_or_none()
-    if row is None:
-        # No admin yet ? leave ownership null; first admin is responsible for
-        # claiming via the UI once they exist.
-        admin_id: str | None = None
-    else:
-        admin_id = str(row.id)
-
-    config = getattr(app.state, "config", None)
-    storage = get_or_new_skill_storage(app_config=config)
-    try:
-        skills = storage.load_skills(enabled_only=False)
-    except Exception:
-        logger.exception("_sync_legacy_skills: could not list on-disk skills")
-        return
-
-    inserted = 0
-    for skill in skills:
-        if skill.category != SkillCategory.CUSTOM:
-            continue
-        try:
-            existing = await skill_store.get_any(skill.name)
-            if existing is not None:
-                continue
-            await skill_store.ensure_legacy(
-                {"name": skill.name, "owner_user_id": admin_id, "published": True}
-            )
-            inserted += 1
-        except Exception:
-            logger.warning("_sync_legacy_skills: could not register %s", skill.name, exc_info=True)
-    if inserted:
-        logger.info("Skill ownership migration: registered %d legacy custom skill(s)", inserted)
-
-
-async def _migrate_orphaned_threads(store, admin_user_id: str) -> int:
-    """Migrate LangGraph store threads with no user_id to the given admin.
-
-    Uses cursor pagination so all orphans are migrated regardless of
-    count. Returns the number of rows migrated.
-    """
-    migrated = 0
-    async for item in _iter_store_items(store, ("threads",)):
-        metadata = item.value.get("metadata", {})
-        if not metadata.get("user_id"):
-            metadata["user_id"] = admin_user_id
-            item.value["metadata"] = metadata
-            await store.aput(("threads",), item.key, item.value)
-            migrated += 1
-    return migrated
-
-
-async def _run_db_migrations() -> None:
-    """Run Alembic migrations at startup so schema changes are applied automatically."""
-    from pathlib import Path
-
-    from alembic import command
-    from alembic.config import Config
-
-    import deerflow.persistence as _persistence_pkg
-    from deerflow.persistence.engine import get_engine
-
-    engine = get_engine()
-    if engine is None:
-        logger.info("Persistence backend=memory ? skipping Alembic migrations")
-        return
-
-    if os.getenv("DEER_FLOW_SKIP_ALEMBIC", "").lower() in {"1", "true", "yes"}:
-        logger.warning("DEER_FLOW_SKIP_ALEMBIC is set; skipping Alembic migrations")
-        return
-
-    db_url = engine.url.render_as_string(hide_password=False).replace("%", "%%")
-    migrations_dir = str(Path(_persistence_pkg.__file__).parent / "migrations")
-    result_queue: queue.Queue[BaseException | None] = queue.Queue(maxsize=1)
-
-    def _run_sync() -> None:
-        try:
-            cfg = Config()
-            cfg.set_main_option("script_location", migrations_dir)
-            cfg.set_main_option("sqlalchemy.url", db_url)
-            command.upgrade(cfg, "head")
-        except BaseException as exc:
-            result_queue.put(exc)
-        else:
-            result_queue.put(None)
-
-    try:
-        try:
-            timeout_s = float(os.getenv("DEER_FLOW_ALEMBIC_TIMEOUT", "60"))
-        except ValueError:
-            timeout_s = 60.0
-
-        worker = threading.Thread(target=_run_sync, name="deerflow-alembic-upgrade", daemon=True)
-        worker.start()
-        try:
-            deadline = asyncio.get_running_loop().time() + timeout_s
-            while result_queue.empty():
-                if asyncio.get_running_loop().time() >= deadline:
-                    raise asyncio.TimeoutError
-                await asyncio.sleep(0.2)
-            result = result_queue.get_nowait()
-        except asyncio.TimeoutError:
-            logger.error(
-                "Database migration did not finish within %.0fs; continuing startup on create_all schema. "
-                "Likely cause: MySQL metadata-lock wait or a slow DDL. Set DEER_FLOW_SKIP_ALEMBIC=1 "
-                "to skip startup migrations, or inspect MySQL with SHOW FULL PROCESSLIST.",
-                timeout_s,
-            )
-            return
-        if result is not None:
-            raise result
-        logger.info("Database migrations applied (head)")
-    except Exception:
-        logger.exception("Database migration failed ? schema may be out of date")
-
-
-# Skill curator ?????
-# - ????:?????????"??????";??????? should_run_now() ???
-# - ????:?????????????,???????
-_CURATOR_CHECK_INTERVAL_SECONDS = 3600  # 1 ??
-_CURATOR_STARTUP_DELAY_SECONDS = 300  # 5 ??
-
-
-def _acquire_singleton_lock(lock_path):
-    """???????????,??? worker ?? leader ???
-
-    ?? -> ??????????(?????????,??????/????);
-    ????(???? worker ??)????? -> ?? None?
-    ??????? OS ????????
-    """
-    try:
-        lock_path.parent.mkdir(parents=True, exist_ok=True)
-        handle = open(lock_path, "a+")
-    except OSError:
-        logger.exception("Curator scheduler: ???????")
-        return None
-    try:
-        if sys.platform == "win32":
-            import msvcrt
-
-            handle.seek(0)
-            msvcrt.locking(handle.fileno(), msvcrt.LK_NBLCK, 1)
-        else:
-            import fcntl
-
-            fcntl.flock(handle.fileno(), fcntl.LOCK_EX | fcntl.LOCK_NB)
-    except OSError:
-        handle.close()
-        return None
-    return handle
-
-
-def _release_singleton_lock(handle) -> None:
-    """????? _acquire_singleton_lock ?????????"""
-    try:
-        if sys.platform == "win32":
-            import msvcrt
-
-            handle.seek(0)
-            msvcrt.locking(handle.fileno(), msvcrt.LK_UNLCK, 1)
-        else:
-            import fcntl
-
-            fcntl.flock(handle.fileno(), fcntl.LOCK_UN)
-    except OSError:
-        pass
-    finally:
-        try:
-            handle.close()
-        except OSError:
-            pass
-
-
-async def _curator_scheduler_loop() -> None:
-    """??????:?? _CURATOR_CHECK_INTERVAL_SECONDS ???????????
-
-    ? worker ???????? leader ?? ?? ???????? worker ????
-    ????,?? worker ????,?? curator ?????????? LLM?
-    ? cancel ? CancelledError ??? ``except Exception``(?? BaseException
-    ??,?????),?????????? finally ?????
-    """
-    from deerflow.config.runtime_paths import runtime_home
-    from deerflow.skills.curator import maybe_run_curator
-
-    lock_handle = _acquire_singleton_lock(runtime_home() / ".curator_scheduler.lock")
-    if lock_handle is None:
-        logger.info("Curator scheduler: ??? worker ??????,? worker ??")
-        return
-
-    logger.info("Curator scheduler: ?????,? worker ?? curator ??")
-    try:
-        # ?????????????,??????
-        await asyncio.sleep(_CURATOR_STARTUP_DELAY_SECONDS)
-        while True:
-            try:
-                await maybe_run_curator()
-            except Exception:
-                logger.exception("Curator scheduler tick failed (non-fatal)")
-            await asyncio.sleep(_CURATOR_CHECK_INTERVAL_SECONDS)
-    finally:
-        _release_singleton_lock(lock_handle)
-
-
-def _env_bool(name: str, default: bool) -> bool:
-    raw = os.getenv(name)
-    if raw is None:
-        return default
-    return raw.strip().lower() not in {"0", "false", "no", "off"}
-
-
-def _env_int(name: str, default: int, *, alias: str | None = None, minimum: int = 1) -> int:
-    raw = os.getenv(name)
-    if raw is None and alias:
-        raw = os.getenv(alias)
-    if raw is None:
-        raw = str(default)
-    try:
-        value = int(raw)
-    except ValueError:
-        return default
-    return max(minimum, value)
-
-
-def _env_float(name: str, default: float, *, alias: str | None = None, minimum: float = 0.0) -> float:
-    raw = os.getenv(name)
-    if raw is None and alias:
-        raw = os.getenv(alias)
-    if raw is None:
-        raw = str(default)
-    try:
-        value = float(raw)
-    except ValueError:
-        return default
-    return max(minimum, value)
-
-
-_LEADERBOARD_SNAPSHOT_BACKGROUND_ENABLED = _env_bool("ADMIN_LEADERBOARD_BACKGROUND_ENABLED", True)
-_LEADERBOARD_SNAPSHOT_PREWARM_DAYS = _env_int(
-    "ADMIN_LEADERBOARD_BACKGROUND_PREWARM_DAYS",
-    30,
-)
-_LEADERBOARD_SNAPSHOT_BATCH_SIZE = _env_int(
-    "ADMIN_LEADERBOARD_BACKGROUND_BATCH_SIZE",
-    1,
-)
-_LEADERBOARD_SNAPSHOT_INTERVAL_SECONDS = _env_float(
-    "ADMIN_LEADERBOARD_BACKGROUND_INTERVAL_SECONDS",
-    300.0,
-)
-_LEADERBOARD_SNAPSHOT_SLEEP_SECONDS = _env_float(
-    "ADMIN_LEADERBOARD_BACKGROUND_SLEEP_SECONDS",
-    2.0,
-)
-_LEADERBOARD_SNAPSHOT_RUNNING_TTL_SECONDS = _env_int(
-    "ADMIN_LEADERBOARD_BACKGROUND_RUNNING_TTL_SECONDS",
-    1800,
-)
-_LEADERBOARD_SNAPSHOT_STARTUP_DELAY_SECONDS = _env_float(
-    "ADMIN_LEADERBOARD_BACKGROUND_STARTUP_DELAY_SECONDS",
-    60.0,
-)
-# Concurrency (system-pressure) sampler: how often to snapshot the in-flight run
-# count, and how long to keep samples (drives the admin 并发量 line chart, which
-# aggregates per minute/hour). Sampling is in-memory cheap (one tiny INSERT).
-_CONCURRENCY_SAMPLE_ENABLED = _env_bool("CONCURRENCY_SAMPLE_ENABLED", True)
-_CONCURRENCY_SAMPLE_INTERVAL_SECONDS = _env_float(
-    "CONCURRENCY_SAMPLE_INTERVAL_SECONDS",
-    15.0,
-)
-_CONCURRENCY_SAMPLE_RETENTION_HOURS = _env_float(
-    "CONCURRENCY_SAMPLE_RETENTION_HOURS",
-    168.0,  # 7 days
-)
-# Hour (local Beijing time, 0-23) after which the scheduler force-recomputes the
-# *previous* day's snapshot once, so records that landed near midnight are fully
-# captured. Default 1 ? "???? 1 ????????".
-_LEADERBOARD_FINALIZE_HOUR = min(
-    23,
-    _env_int("ADMIN_LEADERBOARD_FINALIZE_HOUR", 1, minimum=0),
-)
-
-
-def _leaderboard_scheduler_sql_for_log(stmt) -> str:
-    try:
-        return str(stmt.compile(compile_kwargs={"literal_binds": True}))
-    except Exception:
-        return str(stmt)
-
-
-async def _execute_leaderboard_scheduler_query(session, step: str, stmt, **context):
-    sql = _leaderboard_scheduler_sql_for_log(stmt)
-    logger.info(
-        "Leaderboard snapshot DB query start: step=%s stat_date=%s job=%s settings_hash=%s started_at=%s sql=%s",
-        step,
-        context.get("stat_date") or context.get("stat_dates") or "",
-        context.get("job_key") or "",
-        context.get("settings_hash") or "",
-        datetime.now().isoformat(),
-        sql,
-    )
-    start = perf_counter()
-    try:
-        result = await session.execute(stmt)
-    except Exception:
-        logger.exception(
-            "Leaderboard snapshot DB query failed: step=%s stat_date=%s job=%s settings_hash=%s elapsed_ms=%.2f",
-            step,
-            context.get("stat_date") or context.get("stat_dates") or "",
-            context.get("job_key") or "",
-            context.get("settings_hash") or "",
-            (perf_counter() - start) * 1000,
-        )
-        raise
-    logger.info(
-        "Leaderboard snapshot DB query done: step=%s stat_date=%s job=%s settings_hash=%s finished_at=%s elapsed_ms=%.2f",
-        step,
-        context.get("stat_date") or context.get("stat_dates") or "",
-        context.get("job_key") or "",
-        context.get("settings_hash") or "",
-        datetime.now().isoformat(),
-        (perf_counter() - start) * 1000,
-    )
-    return result
-
-
-async def _time_leaderboard_scheduler_db_operation(step: str, operation, *, sql: str = "", **context):
-    logger.info(
-        "Leaderboard snapshot DB query start: step=%s stat_date=%s job=%s settings_hash=%s started_at=%s sql=%s",
-        step,
-        context.get("stat_date") or context.get("stat_dates") or "",
-        context.get("job_key") or "",
-        context.get("settings_hash") or "",
-        datetime.now().isoformat(),
-        sql,
-    )
-    start = perf_counter()
-    try:
-        result = await operation()
-    except Exception:
-        logger.exception(
-            "Leaderboard snapshot DB query failed: step=%s stat_date=%s job=%s settings_hash=%s elapsed_ms=%.2f",
-            step,
-            context.get("stat_date") or context.get("stat_dates") or "",
-            context.get("job_key") or "",
-            context.get("settings_hash") or "",
-            (perf_counter() - start) * 1000,
-        )
-        raise
-    logger.info(
-        "Leaderboard snapshot DB query done: step=%s stat_date=%s job=%s settings_hash=%s finished_at=%s elapsed_ms=%.2f",
-        step,
-        context.get("stat_date") or context.get("stat_dates") or "",
-        context.get("job_key") or "",
-        context.get("settings_hash") or "",
-        datetime.now().isoformat(),
-        (perf_counter() - start) * 1000,
-    )
-    return result
-
-
-def _recent_leaderboard_stat_dates(days: int) -> list[str]:
-    from datetime import datetime, timedelta
-
-    from deerflow.persistence.types import BEIJING_TZ
-
-    today = datetime.now(BEIJING_TZ).date()
-    start = today - timedelta(days=max(1, days) - 1)
-    return [(start + timedelta(days=i)).isoformat() for i in range((today - start).days + 1)]
-
-
-async def _leaderboard_snapshot_due_dates(
-    session_factory,
-    *,
-    settings_hash: str,
-    limit: int,
-    running_ttl_seconds: int,
-) -> list[str]:
-    from datetime import UTC, datetime, timedelta
-
-    from sqlalchemy import and_, or_, select
-
-    from deerflow.persistence.admin_stats.model import AdminLeaderboardDailyStatRow
-
-    cutoff = datetime.now(UTC) - timedelta(seconds=running_ttl_seconds)
-    async with session_factory() as session:
-        stmt = (
-            select(AdminLeaderboardDailyStatRow.stat_date)
-            .where(
-                AdminLeaderboardDailyStatRow.settings_hash == settings_hash,
-                or_(
-                    AdminLeaderboardDailyStatRow.status.in_(["queued", "error"]),
-                    and_(
-                        AdminLeaderboardDailyStatRow.status == "running",
-                        AdminLeaderboardDailyStatRow.updated_at < cutoff,
-                    ),
-                ),
-            )
-            .order_by(AdminLeaderboardDailyStatRow.stat_date.desc())
-            .limit(limit)
-        )
-        rows = (
-            await _execute_leaderboard_scheduler_query(
-                session,
-                "scheduler.due_dates",
-                stmt,
-                settings_hash=settings_hash,
-            )
-        ).all()
-    return [str(row[0]) for row in rows]
-
-
-async def _claim_leaderboard_snapshot(
-    session_factory,
-    *,
-    stat_date: str,
-    settings_hash: str,
-    running_ttl_seconds: int,
-) -> bool:
-    from datetime import UTC, datetime
-
-    from app.gateway.routers.admin_users import _snapshot_is_fresh
-    from deerflow.persistence.admin_stats.model import AdminLeaderboardDailyStatRow
-    from deerflow.persistence.types import BEIJING_TZ
-
-    async with session_factory() as session:
-        row = await _time_leaderboard_scheduler_db_operation(
-            "scheduler.claim.get",
-            lambda: session.get(
-                AdminLeaderboardDailyStatRow,
-                {"stat_date": stat_date, "settings_hash": settings_hash},
-            ),
-            sql="SELECT admin_leaderboard_daily_stats by primary key",
-            stat_date=stat_date,
-            settings_hash=settings_hash,
-        )
-        if row is None or _snapshot_is_fresh(row, stat_date):
-            return False
-        if row.status == "running":
-            updated_at = row.updated_at
-            if isinstance(updated_at, datetime):
-                if updated_at.tzinfo is None:
-                    updated_at = updated_at.replace(tzinfo=BEIJING_TZ)
-                age = (datetime.now(BEIJING_TZ) - updated_at.astimezone(BEIJING_TZ)).total_seconds()
-                if age < running_ttl_seconds:
-                    return False
-        row.status = "running"
-        row.message = "?????????????"
-        row.error = None
-        row.updated_at = datetime.now(UTC)
-        await _time_leaderboard_scheduler_db_operation(
-            "scheduler.claim.commit",
-            lambda: session.commit(),
-            sql="COMMIT",
-            stat_date=stat_date,
-            settings_hash=settings_hash,
-        )
-    return True
-
-
-async def _maybe_finalize_previous_leaderboard_day(session_factory, settings_hash: str) -> bool:
-    """Once per day after ``_LEADERBOARD_FINALIZE_HOUR``, force-recompute yesterday.
-
-    A ``ready`` past-day snapshot is otherwise treated as fresh forever and never
-    rebuilt. But the previous day's snapshot is usually generated before midnight,
-    so it can miss runs/tool-calls that only land in the DB around the day
-    boundary. This re-queues yesterday (``force=True``) a single time; the rebuild
-    stamps ``generated_at`` to today, which stops it from re-firing for the rest of
-    the day.
-    """
-    from datetime import timedelta
-
-    from app.gateway.routers.admin_users import (
-        _as_beijing,
-        _queue_daily_leaderboard_snapshots,
-    )
-    from deerflow.persistence.admin_stats.finalize import should_finalize_previous_day
-    from deerflow.persistence.admin_stats.model import AdminLeaderboardDailyStatRow
-    from deerflow.persistence.types import BEIJING_TZ
-
-    now = datetime.now(BEIJING_TZ)
-    yesterday = (now.date() - timedelta(days=1)).isoformat()
-    async with session_factory() as session:
-        row = await session.get(
-            AdminLeaderboardDailyStatRow,
-            {"stat_date": yesterday, "settings_hash": settings_hash},
-        )
-    generated_at = _as_beijing(row.generated_at) if (row is not None and row.generated_at is not None) else None
-    if not should_finalize_previous_day(
-        status=(row.status if row is not None else None),
-        generated_at=generated_at,
-        now=now,
-        finalize_hour=_LEADERBOARD_FINALIZE_HOUR,
-    ):
-        return False
-    queued = await _queue_daily_leaderboard_snapshots(
-        session_factory,
-        [yesterday],
-        settings_hash,
-        force=True,
-        message="???????????????????????????",
-    )
-    if queued:
-        logger.info(
-            "Leaderboard snapshot finalize: queued previous day %s for full recompute (hash=%s)",
-            yesterday,
-            settings_hash[:10],
-        )
-        return True
-    return False
-
-
-async def _run_leaderboard_snapshot_tick() -> tuple[int, int]:
-    from app.gateway.routers.admin_users import (
-        _daily_job_key,
-        _queue_daily_leaderboard_snapshots,
-        _refresh_daily_leaderboard_snapshot,
-        _resolve_leaderboard_snapshot_scope,
-    )
-    from deerflow.config.system_settings import load_system_settings
-    from deerflow.persistence.engine import get_session_factory
-
-    session_factory = get_session_factory()
-    if session_factory is None:
-        return 0, 0
-
-    settings = load_system_settings().leaderboard
-    excluded, include_scheduled, include_admins, include_failed, settings_hash = _resolve_leaderboard_snapshot_scope(settings)
-    stat_dates = _recent_leaderboard_stat_dates(_LEADERBOARD_SNAPSHOT_PREWARM_DAYS)
-    queued_dates = await _queue_daily_leaderboard_snapshots(
-        session_factory,
-        stat_dates,
-        settings_hash,
-        message="?????????????/?????",
-    )
-
-    # ??????? 1 ???????"???"????????????????????
-    try:
-        await _maybe_finalize_previous_leaderboard_day(session_factory, settings_hash)
-    except Exception:
-        logger.exception("Leaderboard snapshot finalize check failed (non-fatal)")
-
-    due_dates = await _leaderboard_snapshot_due_dates(
-        session_factory,
-        settings_hash=settings_hash,
-        limit=_LEADERBOARD_SNAPSHOT_BATCH_SIZE,
-        running_ttl_seconds=_LEADERBOARD_SNAPSHOT_RUNNING_TTL_SECONDS,
-    )
-    processed = 0
-    for stat_date in due_dates:
-        claimed = await _claim_leaderboard_snapshot(
-            session_factory,
-            stat_date=stat_date,
-            settings_hash=settings_hash,
-            running_ttl_seconds=_LEADERBOARD_SNAPSHOT_RUNNING_TTL_SECONDS,
-        )
-        if not claimed:
-            continue
-        logger.info("Leaderboard snapshot scheduler: building date=%s hash=%s", stat_date, settings_hash[:10])
-        await _refresh_daily_leaderboard_snapshot(
-            _daily_job_key(stat_date, settings_hash),
-            session_factory=session_factory,
-            settings=settings,
-            stat_date=stat_date,
-            settings_hash=settings_hash,
-            excluded=excluded,
-            include_scheduled=include_scheduled,
-            include_admins=include_admins,
-            include_failed=include_failed,
-        )
-        processed += 1
-        if _LEADERBOARD_SNAPSHOT_SLEEP_SECONDS > 0 and processed < len(due_dates):
-            await asyncio.sleep(_LEADERBOARD_SNAPSHOT_SLEEP_SECONDS)
-    return processed, len(queued_dates)
-
-
-async def _leaderboard_snapshot_scheduler_loop() -> None:
-    """Background task inside the existing gateway process for daily snapshots."""
-    if not _LEADERBOARD_SNAPSHOT_BACKGROUND_ENABLED:
-        logger.info("Leaderboard snapshot scheduler disabled by ADMIN_LEADERBOARD_BACKGROUND_ENABLED")
-        return
-
-    from deerflow.config.runtime_paths import runtime_home
-
-    lock_handle = _acquire_singleton_lock(runtime_home() / ".leaderboard_snapshot_scheduler.lock")
-    if lock_handle is None:
-        logger.info("Leaderboard snapshot scheduler: another gateway worker holds the lock; this worker skips")
-        return
-
-    logger.info(
-        "Leaderboard snapshot scheduler started: prewarm_days=%d batch_size=%d interval=%.1fs",
-        _LEADERBOARD_SNAPSHOT_PREWARM_DAYS,
-        _LEADERBOARD_SNAPSHOT_BATCH_SIZE,
-        _LEADERBOARD_SNAPSHOT_INTERVAL_SECONDS,
-    )
-    try:
-        if _LEADERBOARD_SNAPSHOT_STARTUP_DELAY_SECONDS > 0:
-            await asyncio.sleep(_LEADERBOARD_SNAPSHOT_STARTUP_DELAY_SECONDS)
-        while True:
-            try:
-                processed, queued = await _run_leaderboard_snapshot_tick()
-                if processed or queued:
-                    logger.info(
-                        "Leaderboard snapshot scheduler tick: queued=%d processed=%d",
-                        queued,
-                        processed,
-                    )
-            except Exception:
-                logger.exception("Leaderboard snapshot scheduler tick failed (non-fatal)")
-            await asyncio.sleep(_LEADERBOARD_SNAPSHOT_INTERVAL_SECONDS)
-    finally:
-        _release_singleton_lock(lock_handle)
-
-
-async def _concurrency_sampler_loop(app: FastAPI) -> None:
-    """Periodically record the in-flight run count for the admin pressure chart.
-
-    Runs in the gateway process, singleton-locked so only one worker samples.
-    Each tick records ``len(run_manager.list_active())`` and, roughly hourly,
-    purges samples older than the retention window. Best-effort throughout — a
-    failed tick is logged and the loop continues.
-    """
-    if not _CONCURRENCY_SAMPLE_ENABLED:
-        logger.info("Concurrency sampler disabled by CONCURRENCY_SAMPLE_ENABLED")
-        return
-
-    store = getattr(app.state, "concurrency_sample_store", None)
-    run_mgr = getattr(app.state, "run_manager", None)
-    concurrency_gate = getattr(app.state, "concurrency_gate", None)
-    if store is None or run_mgr is None:
-        logger.info("Concurrency sampler not started (no store / run manager — memory backend?)")
-        return
-
-    from deerflow.config.runtime_paths import runtime_home
-
-    lock_handle = _acquire_singleton_lock(runtime_home() / ".concurrency_sampler.lock")
-    if lock_handle is None:
-        logger.info("Concurrency sampler: another gateway worker holds the lock; this worker skips")
-        return
-
-    interval = max(1.0, _CONCURRENCY_SAMPLE_INTERVAL_SECONDS)
-    purge_every_ticks = max(1, int(3600 / interval))  # ~once per hour
-    logger.info("Concurrency sampler started: interval=%.1fs retention=%.1fh", interval, _CONCURRENCY_SAMPLE_RETENTION_HOURS)
-    ticks = 0
-    try:
-        while True:
-            try:
-                # Runtime toggle (admin can turn sampling off without a restart).
-                # mtime-cached read, so this is cheap every tick.
-                from deerflow.config.system_settings import get_concurrency_monitor_settings
-
-                if get_concurrency_monitor_settings().enabled:
-                    active_count = None
-                    if concurrency_gate is not None:
-                        active_count = await concurrency_gate.count_active()
-                    if active_count is None:
-                        active = await run_mgr.list_active()
-                        active_count = len(active)
-                    await store.record(active_count)
-                    ticks += 1
-                    if ticks % purge_every_ticks == 0:
-                        from datetime import UTC, datetime, timedelta
-
-                        cutoff = datetime.now(UTC) - timedelta(hours=_CONCURRENCY_SAMPLE_RETENTION_HOURS)
-                        await store.purge_older_than(cutoff)
-            except Exception:
-                logger.exception("Concurrency sampler tick failed (non-fatal)")
-            await asyncio.sleep(interval)
-    finally:
-        _release_singleton_lock(lock_handle)
-
-
-def _warn_if_multi_worker() -> None:
-    """?? uvicorn / gunicorn ? worker ???? ERROR ???
-
-    ? agent / canvas / AI ??????? Gateway ????? ``StreamBridge``
-    ?? run_id ? SSE ?????? map?? worker ????? worker A ??
-    ``/runs/stream``???? worker B ? join/resume??? in-process map ??
-    ???? 404 / ???
-
-    ????? startup ? ERROR ????**???? fail-fast** ?? ??
-    ???????????????? sticky session??????????
-    """
-    # uvicorn / gunicorn ??????? worker ?????????????
-    # ?????? sibling ??????????? WEB_CONCURRENCY?gunicorn
-    # / uvicorn ????
-    web_concurrency = os.environ.get("WEB_CONCURRENCY", "").strip()
-    if web_concurrency and web_concurrency.isdigit() and int(web_concurrency) > 1:
-        logger.error(
-            "? ??? WEB_CONCURRENCY=%s?? worker ????AI ?? / ? agent "
-            "??????????????? worker ????????? / "
-            "SSE ?? / resume 400??????????? WEB_CONCURRENCY=1 "
-            "????????? sticky session??? AIWritingPage.tsx ????",
-            web_concurrency,
-        )
-        return
-
-    # ???? /proc ??? uvicorn ?????? Linux ??????
-    if not sys.platform.startswith("linux"):
-        return
-    try:
-        from pathlib import Path
-
-        peers = 0
-        for entry in Path("/proc").iterdir():
-            if not entry.name.isdigit():
-                continue
-            try:
-                cmdline = (entry / "cmdline").read_bytes().replace(b"\x00", b" ").decode("utf-8", "ignore")
-            except OSError:
-                continue
-            if "uvicorn" in cmdline and "app.gateway.app" in cmdline:
-                peers += 1
-        if peers > 1:
-            logger.error(
-                "? ??? %d ??? uvicorn ???? worker ????? AI ?? "
-                "????? / SSE ??????????? worker ??????? "
-                "?? sticky session?",
-                peers,
-            )
-    except Exception:
-        # ????????? ?? ??????
-        pass
-
-
-@asynccontextmanager
-async def lifespan(app: FastAPI) -> AsyncGenerator[None, None]:
-    """Application lifespan handler."""
-
-    # ?????? worker ?? ?? ?? AI ?? / ? agent ???????
-    # ??????? worker ???????? ERROR ?????
-    _warn_if_multi_worker()
-
-    # Load config and check necessary environment variables at startup
-    try:
-        app.state.config = get_app_config()
-        apply_logging_level(app.state.config.log_level)
-        logger.info("Configuration loaded successfully")
-    except Exception as e:
-        error_msg = f"Failed to load configuration during gateway startup: {e}"
-        logger.exception(error_msg)
-        raise RuntimeError(error_msg) from e
-    config = get_gateway_config()
-    logger.info(f"Starting API Gateway on {config.host}:{config.port}")
-
-    # Initialize LangGraph runtime components (StreamBridge, RunManager, checkpointer, store)
-    async with langgraph_runtime(app):
-        logger.info("LangGraph runtime initialised")
-
-        # Run Alembic migrations before anything else touches the schema.
-        await _run_db_migrations()
-
-        # ?? recall_budget cache?? DB ????,? get_effective_memory_config ?????
-        try:
-            from deerflow.config.memory_config import load_recall_budget_cache
-
-            _mc_store = getattr(app.state, "memory_config_store", None)
-            if _mc_store is not None:
-                _recall_budgets = await _mc_store.load_all()
-                load_recall_budget_cache(_recall_budgets)
-                logger.info("Memory config cache warmed: %d user override(s)", len(_recall_budgets))
-        except Exception:
-            logger.exception("Memory config cache warm-up failed (non-fatal)")
-
-        # Ensure admin user exists (auto-create on first boot)
-        # Must run AFTER langgraph_runtime so app.state.store is available for thread migration
-        await _ensure_admin_user(app)
-
-        # Migrate "ownerless" on-disk skills into the new SkillStore table ?
-        # makes the multitenancy switch transparent for existing deployments.
-        try:
-            await _sync_legacy_skills(app)
-        except Exception:
-            logger.exception("Skill ownership migration failed (non-fatal)")
-
-        # ????? agent(intent / recommender / coordinator)???????
-        # ??? _sync_legacy_agents **??**??????????????????
-        # ??? upsert ? agents ?,?????? /api/agents ?????,
-        # ????????????????.deer-flow/ ? gitignore,??? clone
-        # ????????,seeder ????? _roundtable_seed_assets/ ??????
-        try:
-            from app.gateway.routers._roundtable_seed import (
-                ensure_roundtable_functional_agents,
-            )
-
-            ensure_roundtable_functional_agents()
-        except Exception:
-            logger.exception(
-                "Roundtable functional agent seeding failed at startup (non-fatal)"
-            )
-
-        # AI ????? agent(researcher / outliner / writer / editor)???????
-        # ???:??? _sync_legacy_agents ??,?????????????
-        # upsert ? agents ?,admin ?????????????????? agent?
-        try:
-            from app.gateway.routers._ai_writing_seed import (
-                ensure_ai_writing_functional_agents,
-            )
-
-            ensure_ai_writing_functional_agents()
-        except Exception:
-            logger.exception(
-                "AI-writing functional agent seeding failed at startup (non-fatal)"
-            )
-
-        # 内置「模板数据转换器」agent(JSON → HTML 模板填充)。同样必须在
-        # _sync_legacy_agents **之前**补建目录,后者把它 upsert 进 agents 表,
-        # 第一次启动 /api/agents 与定时任务下拉框里就有它。
-        try:
-            from app.gateway.routers._template_builder_seed import (
-                ensure_template_builder_agent,
-            )
-
-            ensure_template_builder_agent()
-        except Exception:
-            logger.exception(
-                "Template-builder agent seeding failed at startup (non-fatal)"
-            )
-
-        # 深度研究「收集员」内置 agent(对话式信息收集)。同样必须在
-        # _sync_legacy_agents 之前补建目录,由后者 upsert 进 agents 表,
-        # 深度研究页的收集对话与智能体管理页即可直接使用。
-        try:
-            from app.gateway.routers._deep_research_seed import (
-                ensure_deep_research_collector_agent,
-            )
-
-            ensure_deep_research_collector_agent()
-        except Exception:
-            logger.exception(
-                "Deep-research collector agent seeding failed at startup (non-fatal)"
-            )
-
-        # ?????????????????3Q/6BF/7BF/8BF??agent/chain ?????
-        try:
-            business_mapping_store = getattr(app.state, "business_mapping_store", None)
-            if business_mapping_store is not None:
-                inserted = await business_mapping_store.ensure_seed_defaults()
-                if inserted:
-                    logger.info("Seeded %d business mapping rows", inserted)
-        except Exception:
-            logger.exception("Business mapping seeding failed at startup (non-fatal)")
-
-        # ???? .deer-flow/agents/{id}/ ???????? agent ?? agents ??
-        # ??????????????? sync(????? + N ? agent ??? DB
-        # SELECT/UPDATE),??????? + ?????? TTL ??,????
-        # /api/agents ?????
-        try:
-            from app.gateway.routers.agents import (
-                _mark_legacy_sync_fresh,
-                _sync_legacy_agents,
-            )
-            from app.gateway.services import _mark_legacy_sync_fresh as _services_mark_sync_fresh
-
-            agent_store = getattr(app.state, "agent_store", None)
-            if agent_store is not None:
-                await _sync_legacy_agents(agent_store)
-                _mark_legacy_sync_fresh()
-                # ``services.py`` keeps an independent TTL gate (used by
-                # ``start_run`` ? ``_enforce_agent_access``). Without this
-                # ack, the first ``/runs/stream`` after startup would
-                # re-run the heavy sync even though we just did it.
-                _services_mark_sync_fresh()
-                logger.info("Legacy folder-backed agents synced at startup")
-                # 内置功能型 agent(roundtable-* / ai-writing-*)按用途默认打标签,
-                # 供管理员「内置智能体」筛选按用途归类。幂等、只追加、失败不阻塞启动。
-                try:
-                    from app.gateway.routers._builtin_agent_tag_seed import (
-                        ensure_builtin_agent_purpose_tags,
-                    )
-
-                    tag_store = getattr(app.state, "tag_store", None)
-                    if tag_store is not None:
-                        await ensure_builtin_agent_purpose_tags(agent_store, tag_store)
-                except Exception:
-                    logger.exception("Built-in agent purpose-tag seeding failed at startup (non-fatal)")
-        except Exception:
-            logger.exception("Legacy agent sync failed at startup (non-fatal)")
-
-        # ?? skill curator ???:??????,?????????????
-        # (per-user interval_hours ??????????????)?
-        try:
-            app.state.curator_scheduler_task = asyncio.create_task(_curator_scheduler_loop())
-        except Exception:
-            logger.exception("Skill curator scheduler failed to start (non-fatal)")
-
-        # ?? AI ????????:? config.yaml ? ai_writing.cleanup ???,
-        # ?????? 3 ????,? retention_days(?? 7 ?)?????,??
-        # ?? MySQL ??? + SQLite checkpointer ?? thread ???? worker
-        # ?????? leader,?????? worker ??????
-        try:
-            from app.gateway.ai_writing_cleanup import cleanup_scheduler_loop
-
-            app.state.ai_writing_cleanup_task = asyncio.create_task(cleanup_scheduler_loop(app))
-        except Exception:
-            logger.exception("AI writing cleanup scheduler failed to start (non-fatal)")
-
-        try:
-            app.state.leaderboard_snapshot_scheduler_task = asyncio.create_task(_leaderboard_snapshot_scheduler_loop())
-        except Exception:
-            logger.exception("Leaderboard snapshot scheduler failed to start (non-fatal)")
-
-        # Concurrency (system-pressure) sampler — lightweight, singleton-locked,
-        # runtime-toggleable via system_settings.concurrency_monitor.enabled.
-        try:
-            app.state.concurrency_sampler_task = asyncio.create_task(_concurrency_sampler_loop(app))
-        except Exception:
-            logger.exception("Concurrency sampler failed to start (non-fatal)")
-
-        # Start IM channel service if any channels are configured
-        try:
-            from app.channels.service import start_channel_service
-
-            channel_service = await start_channel_service(app.state.config)
-            logger.info("Channel service started: %s", channel_service.get_status())
-        except Exception:
-            logger.exception("No IM channels configured or channel service failed to start")
-
-        scheduled_tasks_enabled = os.environ.get("SCHEDULED_TASKS_ENABLED", "true").strip().lower()
-        if scheduled_tasks_enabled in {"0", "false", "no", "off"}:
-            logger.info("Scheduled task service disabled by SCHEDULED_TASKS_ENABLED=%s", scheduled_tasks_enabled)
-        else:
-            try:
-                scheduled_task_store = getattr(app.state, "scheduled_task_store", None)
-                if scheduled_task_store is not None:
-                    scheduler_service = ScheduledTaskService(app, scheduled_task_store)
-                    app.state.scheduled_task_service = scheduler_service
-                    set_scheduled_task_service(scheduler_service)
-                    await scheduler_service.start()
-                    logger.info("Scheduled task service started")
-            except Exception:
-                logger.exception("Scheduled task service failed to start")
-
-        yield
-
-        # ?? curator ???
-        curator_task = getattr(app.state, "curator_scheduler_task", None)
-        if curator_task is not None:
-            curator_task.cancel()
-            try:
-                await curator_task
-            except asyncio.CancelledError:
-                pass
-            except Exception:
-                logger.exception("Curator scheduler shutdown error")
-
-        # ?? AI ???????
-        cleanup_task = getattr(app.state, "ai_writing_cleanup_task", None)
-        if cleanup_task is not None:
-            cleanup_task.cancel()
-            try:
-                await cleanup_task
-            except asyncio.CancelledError:
-                pass
-            except Exception:
-                logger.exception("AI writing cleanup scheduler shutdown error")
-
-        leaderboard_snapshot_task = getattr(app.state, "leaderboard_snapshot_scheduler_task", None)
-        if leaderboard_snapshot_task is not None:
-            leaderboard_snapshot_task.cancel()
-            try:
-                await leaderboard_snapshot_task
-            except asyncio.CancelledError:
-                pass
-            except Exception:
-                logger.exception("Leaderboard snapshot scheduler shutdown error")
-
-        concurrency_sampler_task = getattr(app.state, "concurrency_sampler_task", None)
-        if concurrency_sampler_task is not None:
-            concurrency_sampler_task.cancel()
-            try:
-                await concurrency_sampler_task
-            except asyncio.CancelledError:
-                pass
-            except Exception:
-                logger.exception("Concurrency sampler shutdown error")
-
-        # ?? 7 ????AI ?? graph ???? checkpointer / in-flight task?
-        # thread ??? LangGraph Server ? checkpointer ???lifespan ???
-        # ``langgraph_runtime()`` ? contextmanager ???????????????
-
-        try:
-            scheduler_service = getattr(app.state, "scheduled_task_service", None)
-            if scheduler_service is not None:
-                await scheduler_service.stop()
-        except Exception:
-            logger.exception("Failed to stop scheduled task service")
-        finally:
-            set_scheduled_task_service(None)
-
-        # Stop channel service on shutdown (bounded to prevent worker hang)
-        try:
-            from app.channels.service import stop_channel_service
-
-            await asyncio.wait_for(
-                stop_channel_service(),
-                timeout=_SHUTDOWN_HOOK_TIMEOUT_SECONDS,
-            )
-        except TimeoutError:
-            logger.warning(
-                "Channel service shutdown exceeded %.1fs; proceeding with worker exit.",
-                _SHUTDOWN_HOOK_TIMEOUT_SECONDS,
-            )
-        except Exception:
-            logger.exception("Failed to stop channel service")
-
-    logger.info("Shutting down API Gateway")
-
-
-def create_app() -> FastAPI:
-    """Create and configure the FastAPI application.
-
-    Returns:
-        Configured FastAPI application instance.
-    """
-    config = get_gateway_config()
-    docs_kwargs = {"docs_url": "/docs", "redoc_url": "/redoc", "openapi_url": "/openapi.json"} if config.enable_docs else {"docs_url": None, "redoc_url": None, "openapi_url": None}
-
-    app = FastAPI(
-        title="DeerFlow API Gateway",
-        description="""
-## DeerFlow API Gateway
-
-API Gateway for DeerFlow - A LangGraph-based AI agent backend with sandbox execution capabilities.
-
-### Features
-
-- **Models Management**: Query and retrieve available AI models
-- **MCP Configuration**: Manage Model Context Protocol (MCP) server configurations
-- **Memory Management**: Access and manage global memory data for personalized conversations
-- **Skills Management**: Query and manage skills and their enabled status
-- **Artifacts**: Access thread artifacts and generated files
-- **Health Monitoring**: System health check endpoints
-
-### Architecture
-
-LangGraph requests are handled by nginx reverse proxy.
-This gateway provides custom endpoints for models, MCP configuration, skills, and artifacts.
-        """,
-        version="0.1.0",
-        lifespan=lifespan,
-        **docs_kwargs,
-        openapi_tags=[
-            {
-                "name": "models",
-                "description": "Operations for querying available AI models and their configurations",
-            },
-            {
-                "name": "mcp",
-                "description": "Manage Model Context Protocol (MCP) server configurations",
-            },
-            {
-                "name": "memory",
-                "description": "Access and manage global memory data for personalized conversations",
-            },
-            {
-                "name": "skills",
-                "description": "Manage skills and their configurations",
-            },
-            {
-                "name": "artifacts",
-                "description": "Access and download thread artifacts and generated files",
-            },
-            {
-                "name": "uploads",
-                "description": "Upload and manage user files for threads",
-            },
-            {
-                "name": "threads",
-                "description": "Manage DeerFlow thread-local filesystem data",
-            },
-            {
-                "name": "agents",
-                "description": "Create and manage custom agents with per-agent config and prompts",
-            },
-            {
-                "name": "suggestions",
-                "description": "Generate follow-up question suggestions for conversations",
-            },
-            {
-                "name": "channels",
-                "description": "Manage IM channel integrations (Feishu, Slack, Telegram)",
-            },
-            {
-                "name": "assistants-compat",
-                "description": "LangGraph Platform-compatible assistants API (stub)",
-            },
-            {
-                "name": "runs",
-                "description": "LangGraph Platform-compatible runs lifecycle (create, stream, cancel)",
-            },
-            {
-                "name": "scheduled-tasks",
-                "description": "Create, manage, and execute user scheduled agent tasks",
-            },
-            {
-                "name": "health",
-                "description": "Health check and system status endpoints",
-            },
-        ],
-    )
-
-    # Auth: reject unauthenticated requests to non-public paths (fail-closed safety net)
-    app.add_middleware(AuthMiddleware)
-
-    # CSRF: Double Submit Cookie pattern for state-changing requests
-    app.add_middleware(CSRFMiddleware)
-
-    # CORS
-    if is_auth_disabled():
-        logger.warning(
-            "DEER_FLOW_AUTH_DISABLED is set ? API authentication is bypassed. "
-            "Do not use this in production."
-        )
-
-    def _cors_allow_any_origin() -> bool:
-        """Mirror any browser Origin (works with credentials); cannot use literal '*'."""
-        v = os.environ.get("GATEWAY_CORS_ALLOW_ALL", "").strip().lower()
-        return v in ("1", "true", "yes", "on") or is_auth_disabled()
-
-    # Keep standard local frontend origins available even when the startup
-    # script supplies an explicit production/development allow-list. Vite can
-    # choose 5173 or move to 5174 when that port is already occupied; excluding
-    # the former makes every browser preflight fail after a backend restart.
-    local_vite_origins = [
-        "http://localhost:5173",
-        "http://127.0.0.1:5173",
-        "http://localhost:5174",
-        "http://127.0.0.1:5174",
-        "http://localhost:3000",
-        "http://127.0.0.1:3000",
-    ]
-
-    # The legacy TaskCOP Vue application uses its own Vite development port.
-    taskcop_vite_origins = [
-        "http://localhost:8080",
-        "http://127.0.0.1:8080",
-    ]
-    local_dev_origins = [*local_vite_origins, *taskcop_vite_origins]
-
-    cors_origins_env = os.environ.get("GATEWAY_CORS_ORIGINS", "").strip()
-    cors_origins: list[str] = [o.strip() for o in cors_origins_env.split(",") if o.strip()] if cors_origins_env else []
-    if cors_origins:
-        cors_origins = list(dict.fromkeys([*cors_origins, *local_dev_origins]))
-
-    # Headers the LangGraph SDK / browser clients need to read from cross-origin
-    # responses. Without exposing them, the SDK's `response.headers.get(...)` returns
-    # null on cross-origin streaming responses, breaking run-id extraction and reconnect.
-    cors_expose_headers = [
-        "Content-Disposition",
-        "Content-Location",
-        "Location",
-        "X-Run-Id",
-        "X-Thread-Id",
-    ]
-
-    # Public pages (e.g. http://47.x.x.x:7010) calling this Gateway on localhost
-    # trigger Chrome/Edge Private Network Access preflights. Without
-    # allow_private_network=True the browser reports a generic CORS failure even
-    # when Access-Control-Allow-Origin is correct.
-    cors_kwargs = {
-        "allow_credentials": True,
-        "allow_methods": ["*"],
-        "allow_headers": ["*"],
-        "expose_headers": cors_expose_headers,
-        "allow_private_network": True,
-    }
-
-    if _cors_allow_any_origin():
-        logger.warning(
-            "Permissive CORS enabled ? any Origin is echoed back (GATEWAY_CORS_ALLOW_ALL or "
-            "DEER_FLOW_AUTH_DISABLED). Do not expose publicly."
-        )
-        app.add_middleware(
-            CORSMiddleware,
-            allow_origins=[],
-            # fullmatch() on the Origin header; browsers send scheme://host[:port]
-            allow_origin_regex=r".+",
-            **cors_kwargs,
-        )
-    elif cors_origins:
-        if "*" in cors_origins:
-            logger.error(
-                "GATEWAY_CORS_ORIGINS contains '*' with allow_credentials=True ? invalid. "
-                "Set GATEWAY_CORS_ALLOW_ALL=1 instead, or list explicit origins."
-            )
-            cors_origins = [o for o in cors_origins if o != "*"]
-        if cors_origins:
-            app.add_middleware(
-                CORSMiddleware,
-                allow_origins=cors_origins,
-                **cors_kwargs,
-            )
-    else:
-        # Safe-by-default fallback for local frontend development. Without any CORS
-        # middleware, browser preflight OPTIONS requests will hit routes directly and
-        # fail with 405, which blocks login and all JSON POST APIs.
-        dev_origins = local_dev_origins
-        logger.warning(
-            "No CORS configuration found; enabling local dev origins only: %s. "
-            "Set GATEWAY_CORS_ORIGINS or GATEWAY_CORS_ALLOW_ALL for deployment.",
-            ", ".join(dev_origins),
-        )
-        app.add_middleware(
-            CORSMiddleware,
-            allow_origins=dev_origins,
-            **cors_kwargs,
-        )
-
-    # Include routers
-    # Models API is mounted at /api/models
-    app.include_router(models.router)
-    # Administrator-managed config.yaml model settings and connection tests.
-    app.include_router(model_management.router)
-
-    # MCP API is mounted at /api/mcp
-    app.include_router(mcp.router)
-
-    # DeerFlow-native LLMWiki management/search BFF (conditionally backed by WeKnora).
-    app.include_router(llmwiki.router)
-
-    # Memory API is mounted at /api/memory
-    app.include_router(memory.router)
-
-    # Skills API is mounted at /api/skills
-    app.include_router(skills.router)
-
-    # Artifacts API is mounted at /api/threads/{thread_id}/artifacts
-    app.include_router(artifacts.router)
-
-    # Uploads API is mounted at /api/threads/{thread_id}/uploads
-    app.include_router(uploads.router)
-
-    # Optional Chrome-extension page capture API is mounted at /api/browser-context
-    app.include_router(browser_context.router)
-
-    # Thread cleanup API is mounted at /api/threads/{thread_id}
-    app.include_router(threads.router)
-
-    # Agents API is mounted at /api/agents
-    app.include_router(agents.router)
-
-    # Recommended Questions API is mounted at /api/recommended-questions
-    app.include_router(recommended_questions.router)
-
-    app.include_router(user_prompts.router)
-
-    # Per-user UI preferences at /api/user-preferences
-    app.include_router(user_preferences.router)
-
-    # Roundtable-planning drafts + recommendation history at /api/roundtable-drafts
-    app.include_router(roundtable_drafts.router)
-
-    # Task-scoped roundtable drafts (taskId 深链聊天记录，独立存、不分权) at /api/roundtable-task-drafts
-    app.include_router(roundtable_task_drafts.router)
-
-    # Task-scoped artifact submission status; stores references only, never artifact content.
-    app.include_router(roundtable_artifact_submissions.router)
-
-    # Step3 流程图结构化抽取 + 校验 (SSE) at /api/flow-extract
-    app.include_router(flow_extract.router)
-
-    # Roundtable-planning business chains at /api/roundtable-chains
-    app.include_router(roundtable_chains.router)
-
-    # 岗位协同独立会话 / 节点状态（不使用原圆桌总控或后台作业）
-    # Canonical position-workspace persistence API.
-    app.include_router(position_roundtable.router, prefix="/api")
-    # Compatibility route for intranet gateways whose allowlist only exposes
-    # the established `/api/multi-agent/*` namespace.
-    app.include_router(
-        position_roundtable.router,
-        prefix="/api/multi-agent",
-        include_in_schema=False,
-    )
-
-    # Global business ? agent / chain mapping at /api/business-mapping
-    app.include_router(business_mapping.router)
-
-    # Position-roundtable work-view role configuration at /api/position-roles.
-    app.include_router(position_roles.router)
-
-    # Light-app registry (?????) at /api/light-apps
-    app.include_router(light_apps.router)
-
+    dashboard_sessions,
+    deep_research,
+    html_page_favorites,
+    intent,
+    knowledge,
+    knowledge_ingest,
+    light_apps,
+    llm_metrics,
+    llmwiki,
+    mcp,
+    memory,
+    menu_overrides,
+    model_management,
+    models,
+    multi_agent,
+    notifications,
+    open_3qfx,
+    open_chat,
+    parallel_agent_tasks,
+    parallel_agents,
+    position_roundtable,
+    positions,
+    position_roles,
+    public_config,
+    public_embed,
+    public_leaderboard,
+    public_scheduled_tasks,
+    recommend,
+    flow_extract,
+    recommended_questions,
+    roundtable_chains,
+    roundtable_artifact_submissions,
+    roundtable_diagnostics,
+    roundtable_draft_shares,
+    roundtable_drafts,
+    roundtable_jobs,
+    roundtable_task_drafts,
+    runs,
+    scheduled_tasks,
+    sensitive_words,
+    skill_addresses,
+    skills,
+    suggestions,
+    system_settings,
+    tags,
+    task_buttons,
+    taskcop_tasks,
+    report_structures,
+    task_reports,
+    thread_runs,
+    tool_metrics,
+    thread_shares,
+    threads,
+    uploads,
+    user_preferences,
+    user_prompts,
+    writing,
+)
+from deerflow.config import app_config as deerflow_app_config
+from deerflow.config.app_config import apply_logging_level
+from deerflow.runtime.scheduler import ScheduledTaskService, set_scheduled_task_service
+
+AppConfig = deerflow_app_config.AppConfig
+get_app_config = deerflow_app_config.get_app_config
+
+# Default logging; lifespan overrides from config.yaml log_level.
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
+    datefmt="%Y-%m-%d %H:%M:%S",
+)
+
+logger = logging.getLogger(__name__)
+
+# Upper bound (seconds) each lifespan shutdown hook is allowed to run.
+# Bounds worker exit time so uvicorn's reload supervisor does not keep
+# firing signals into a worker that is stuck waiting for shutdown cleanup.
+_SHUTDOWN_HOOK_TIMEOUT_SECONDS = 5.0
+
+
+async def _ensure_admin_user(app: FastAPI) -> None:
+    """Startup hook: handle first boot and migrate orphan threads otherwise.
+
+    After admin creation, migrate orphan threads from the LangGraph
+    store (metadata.user_id unset) to the admin account. This is the
+    "no-auth ? with-auth" upgrade path: users who ran DeerFlow without
+    authentication have existing LangGraph thread data that needs an
+    owner assigned.
+        First boot (no admin exists):
+            - Does NOT create any user accounts automatically.
+            - The operator must visit ``/setup`` to create the first admin.
+
+    Subsequent boots (admin already exists):
+      - Runs the one-time "no-auth ? with-auth" orphan thread migration for
+        existing LangGraph thread metadata that has no owner_id.
+
+    No SQL persistence migration is needed: the four user_id columns
+    (threads_meta, runs, run_events, feedback) only come into existence
+    alongside the auth module via create_all, so freshly created tables
+    never contain NULL-owner rows.
+    """
+    from sqlalchemy import select
+
+    from app.gateway.deps import get_local_provider
+    from deerflow.persistence.engine import get_session_factory
+    from deerflow.persistence.user.model import UserRow
+
+    try:
+        provider = get_local_provider()
+    except RuntimeError:
+        # Auth persistence may not be initialized in some test/boot paths.
+        # Skip admin migration work rather than failing gateway startup.
+        logger.warning("Auth persistence not ready; skipping admin bootstrap check")
+        return
+
+    sf = get_session_factory()
+    if sf is None:
+        return
+
+    admin_count = await provider.count_admin_users()
+
+    if admin_count == 0:
+        logger.info("=" * 60)
+        logger.info("  First boot detected ? no admin account exists.")
+        logger.info("  Visit /setup to complete admin account creation.")
+        logger.info("=" * 60)
+        return
+
+    # Admin already exists ? run orphan thread migration for any
+    # LangGraph thread metadata that pre-dates the auth module.
+    async with sf() as session:
+        stmt = select(UserRow).where(UserRow.system_role == "admin").limit(1)
+        row = (await session.execute(stmt)).scalar_one_or_none()
+
+    if row is None:
+        return  # Should not happen (admin_count > 0 above), but be safe.
+
+    admin_id = str(row.id)
+
+    # LangGraph store orphan migration ? non-fatal.
+    # This covers the "no-auth ? with-auth" upgrade path for users
+    # whose existing LangGraph thread metadata has no user_id set.
+    store = getattr(app.state, "store", None)
+    if store is not None:
+        try:
+            migrated = await _migrate_orphaned_threads(store, admin_id)
+            if migrated:
+                logger.info("Migrated %d orphan LangGraph thread(s) to admin", migrated)
+        except Exception:
+            logger.exception("LangGraph thread migration failed (non-fatal)")
+
+
+async def _iter_store_items(store, namespace, *, page_size: int = 500):
+    """Paginated async iterator over a LangGraph store namespace.
+
+    Replaces the old hardcoded ``limit=1000`` call with a cursor-style
+    loop so that environments with more than one page of orphans do
+    not silently lose data. Terminates when a page is empty OR when a
+    short page arrives (indicating the last page).
+    """
+    offset = 0
+    while True:
+        batch = await store.asearch(namespace, limit=page_size, offset=offset)
+        if not batch:
+            return
+        for item in batch:
+            yield item
+        if len(batch) < page_size:
+            return
+        offset += page_size
+
+
+async def _sync_legacy_skills(app: FastAPI) -> None:
+    """Idempotently register existing on-disk ``skills/custom/*`` directories
+    in the ``skills`` DB table, owned by the first admin and published=True.
+
+    Without this step, after upgrading to per-user skill ownership the existing
+    custom skill folders would not be visible to anyone (the SkillStore
+    visibility filter relies on a DB row). Running this once on every boot is
+    cheap (folder scan + one ``ensure_legacy`` per name) and idempotent.
+
+    Non-fatal: a failure here MUST NOT block server startup.
+    """
+    from sqlalchemy import select
+
+    from deerflow.persistence.engine import get_session_factory
+    from deerflow.persistence.user.model import UserRow
+    from deerflow.skills.storage import get_or_new_skill_storage
+    from deerflow.skills.types import SkillCategory
+
+    sf = get_session_factory()
+    if sf is None:
+        return
+
+    skill_store = getattr(app.state, "skill_store", None)
+    if skill_store is None:
+        return
+
+    async with sf() as session:
+        stmt = select(UserRow).where(UserRow.system_role == "admin").limit(1)
+        row = (await session.execute(stmt)).scalar_one_or_none()
+    if row is None:
+        # No admin yet ? leave ownership null; first admin is responsible for
+        # claiming via the UI once they exist.
+        admin_id: str | None = None
+    else:
+        admin_id = str(row.id)
+
+    config = getattr(app.state, "config", None)
+    storage = get_or_new_skill_storage(app_config=config)
+    try:
+        skills = storage.load_skills(enabled_only=False)
+    except Exception:
+        logger.exception("_sync_legacy_skills: could not list on-disk skills")
+        return
+
+    inserted = 0
+    for skill in skills:
+        if skill.category != SkillCategory.CUSTOM:
+            continue
+        try:
+            existing = await skill_store.get_any(skill.name)
+            if existing is not None:
+                continue
+            await skill_store.ensure_legacy(
+                {"name": skill.name, "owner_user_id": admin_id, "published": True}
+            )
+            inserted += 1
+        except Exception:
+            logger.warning("_sync_legacy_skills: could not register %s", skill.name, exc_info=True)
+    if inserted:
+        logger.info("Skill ownership migration: registered %d legacy custom skill(s)", inserted)
+
+
+async def _migrate_orphaned_threads(store, admin_user_id: str) -> int:
+    """Migrate LangGraph store threads with no user_id to the given admin.
+
+    Uses cursor pagination so all orphans are migrated regardless of
+    count. Returns the number of rows migrated.
+    """
+    migrated = 0
+    async for item in _iter_store_items(store, ("threads",)):
+        metadata = item.value.get("metadata", {})
+        if not metadata.get("user_id"):
+            metadata["user_id"] = admin_user_id
+            item.value["metadata"] = metadata
+            await store.aput(("threads",), item.key, item.value)
+            migrated += 1
+    return migrated
+
+
+async def _run_db_migrations() -> None:
+    """Run Alembic migrations at startup so schema changes are applied automatically."""
+    from pathlib import Path
+
+    from alembic import command
+    from alembic.config import Config
+
+    import deerflow.persistence as _persistence_pkg
+    from deerflow.persistence.engine import get_engine
+
+    engine = get_engine()
+    if engine is None:
+        logger.info("Persistence backend=memory ? skipping Alembic migrations")
+        return
+
+    if os.getenv("DEER_FLOW_SKIP_ALEMBIC", "").lower() in {"1", "true", "yes"}:
+        logger.warning("DEER_FLOW_SKIP_ALEMBIC is set; skipping Alembic migrations")
+        return
+
+    db_url = engine.url.render_as_string(hide_password=False).replace("%", "%%")
+    migrations_dir = str(Path(_persistence_pkg.__file__).parent / "migrations")
+    result_queue: queue.Queue[BaseException | None] = queue.Queue(maxsize=1)
+
+    def _run_sync() -> None:
+        try:
+            cfg = Config()
+            cfg.set_main_option("script_location", migrations_dir)
+            cfg.set_main_option("sqlalchemy.url", db_url)
+            command.upgrade(cfg, "head")
+        except BaseException as exc:
+            result_queue.put(exc)
+        else:
+            result_queue.put(None)
+
+    try:
+        try:
+            timeout_s = float(os.getenv("DEER_FLOW_ALEMBIC_TIMEOUT", "60"))
+        except ValueError:
+            timeout_s = 60.0
+
+        worker = threading.Thread(target=_run_sync, name="deerflow-alembic-upgrade", daemon=True)
+        worker.start()
+        try:
+            deadline = asyncio.get_running_loop().time() + timeout_s
+            while result_queue.empty():
+                if asyncio.get_running_loop().time() >= deadline:
+                    raise asyncio.TimeoutError
+                await asyncio.sleep(0.2)
+            result = result_queue.get_nowait()
+        except asyncio.TimeoutError:
+            logger.error(
+                "Database migration did not finish within %.0fs; continuing startup on create_all schema. "
+                "Likely cause: MySQL metadata-lock wait or a slow DDL. Set DEER_FLOW_SKIP_ALEMBIC=1 "
+                "to skip startup migrations, or inspect MySQL with SHOW FULL PROCESSLIST.",
+                timeout_s,
+            )
+            return
+        if result is not None:
+            raise result
+        logger.info("Database migrations applied (head)")
+    except Exception:
+        logger.exception("Database migration failed ? schema may be out of date")
+
+
+# Skill curator ?????
+# - ????:?????????"??????";??????? should_run_now() ???
+# - ????:?????????????,???????
+_CURATOR_CHECK_INTERVAL_SECONDS = 3600  # 1 ??
+_CURATOR_STARTUP_DELAY_SECONDS = 300  # 5 ??
+
+
+def _acquire_singleton_lock(lock_path):
+    """???????????,??? worker ?? leader ???
+
+    ?? -> ??????????(?????????,??????/????);
+    ????(???? worker ??)????? -> ?? None?
+    ??????? OS ????????
+    """
+    try:
+        lock_path.parent.mkdir(parents=True, exist_ok=True)
+        handle = open(lock_path, "a+")
+    except OSError:
+        logger.exception("Curator scheduler: ???????")
+        return None
+    try:
+        if sys.platform == "win32":
+            import msvcrt
+
+            handle.seek(0)
+            msvcrt.locking(handle.fileno(), msvcrt.LK_NBLCK, 1)
+        else:
+            import fcntl
+
+            fcntl.flock(handle.fileno(), fcntl.LOCK_EX | fcntl.LOCK_NB)
+    except OSError:
+        handle.close()
+        return None
+    return handle
+
+
+def _release_singleton_lock(handle) -> None:
+    """????? _acquire_singleton_lock ?????????"""
+    try:
+        if sys.platform == "win32":
+            import msvcrt
+
+            handle.seek(0)
+            msvcrt.locking(handle.fileno(), msvcrt.LK_UNLCK, 1)
+        else:
+            import fcntl
+
+            fcntl.flock(handle.fileno(), fcntl.LOCK_UN)
+    except OSError:
+        pass
+    finally:
+        try:
+            handle.close()
+        except OSError:
+            pass
+
+
+async def _curator_scheduler_loop() -> None:
+    """??????:?? _CURATOR_CHECK_INTERVAL_SECONDS ???????????
+
+    ? worker ???????? leader ?? ?? ???????? worker ????
+    ????,?? worker ????,?? curator ?????????? LLM?
+    ? cancel ? CancelledError ??? ``except Exception``(?? BaseException
+    ??,?????),?????????? finally ?????
+    """
+    from deerflow.config.runtime_paths import runtime_home
+    from deerflow.skills.curator import maybe_run_curator
+
+    lock_handle = _acquire_singleton_lock(runtime_home() / ".curator_scheduler.lock")
+    if lock_handle is None:
+        logger.info("Curator scheduler: ??? worker ??????,? worker ??")
+        return
+
+    logger.info("Curator scheduler: ?????,? worker ?? curator ??")
+    try:
+        # ?????????????,??????
+        await asyncio.sleep(_CURATOR_STARTUP_DELAY_SECONDS)
+        while True:
+            try:
+                await maybe_run_curator()
+            except Exception:
+                logger.exception("Curator scheduler tick failed (non-fatal)")
+            await asyncio.sleep(_CURATOR_CHECK_INTERVAL_SECONDS)
+    finally:
+        _release_singleton_lock(lock_handle)
+
+
+def _env_bool(name: str, default: bool) -> bool:
+    raw = os.getenv(name)
+    if raw is None:
+        return default
+    return raw.strip().lower() not in {"0", "false", "no", "off"}
+
+
+def _env_int(name: str, default: int, *, alias: str | None = None, minimum: int = 1) -> int:
+    raw = os.getenv(name)
+    if raw is None and alias:
+        raw = os.getenv(alias)
+    if raw is None:
+        raw = str(default)
+    try:
+        value = int(raw)
+    except ValueError:
+        return default
+    return max(minimum, value)
+
+
+def _env_float(name: str, default: float, *, alias: str | None = None, minimum: float = 0.0) -> float:
+    raw = os.getenv(name)
+    if raw is None and alias:
+        raw = os.getenv(alias)
+    if raw is None:
+        raw = str(default)
+    try:
+        value = float(raw)
+    except ValueError:
+        return default
+    return max(minimum, value)
+
+
+_LEADERBOARD_SNAPSHOT_BACKGROUND_ENABLED = _env_bool("ADMIN_LEADERBOARD_BACKGROUND_ENABLED", True)
+_LEADERBOARD_SNAPSHOT_PREWARM_DAYS = _env_int(
+    "ADMIN_LEADERBOARD_BACKGROUND_PREWARM_DAYS",
+    30,
+)
+_LEADERBOARD_SNAPSHOT_BATCH_SIZE = _env_int(
+    "ADMIN_LEADERBOARD_BACKGROUND_BATCH_SIZE",
+    1,
+)
+_LEADERBOARD_SNAPSHOT_INTERVAL_SECONDS = _env_float(
+    "ADMIN_LEADERBOARD_BACKGROUND_INTERVAL_SECONDS",
+    300.0,
+)
+_LEADERBOARD_SNAPSHOT_SLEEP_SECONDS = _env_float(
+    "ADMIN_LEADERBOARD_BACKGROUND_SLEEP_SECONDS",
+    2.0,
+)
+_LEADERBOARD_SNAPSHOT_RUNNING_TTL_SECONDS = _env_int(
+    "ADMIN_LEADERBOARD_BACKGROUND_RUNNING_TTL_SECONDS",
+    1800,
+)
+_LEADERBOARD_SNAPSHOT_STARTUP_DELAY_SECONDS = _env_float(
+    "ADMIN_LEADERBOARD_BACKGROUND_STARTUP_DELAY_SECONDS",
+    60.0,
+)
+# Concurrency (system-pressure) sampler: how often to snapshot the in-flight run
+# count, and how long to keep samples (drives the admin 并发量 line chart, which
+# aggregates per minute/hour). Sampling is in-memory cheap (one tiny INSERT).
+_CONCURRENCY_SAMPLE_ENABLED = _env_bool("CONCURRENCY_SAMPLE_ENABLED", True)
+_CONCURRENCY_SAMPLE_INTERVAL_SECONDS = _env_float(
+    "CONCURRENCY_SAMPLE_INTERVAL_SECONDS",
+    15.0,
+)
+_CONCURRENCY_SAMPLE_RETENTION_HOURS = _env_float(
+    "CONCURRENCY_SAMPLE_RETENTION_HOURS",
+    168.0,  # 7 days
+)
+# Hour (local Beijing time, 0-23) after which the scheduler force-recomputes the
+# *previous* day's snapshot once, so records that landed near midnight are fully
+# captured. Default 1 ? "???? 1 ????????".
+_LEADERBOARD_FINALIZE_HOUR = min(
+    23,
+    _env_int("ADMIN_LEADERBOARD_FINALIZE_HOUR", 1, minimum=0),
+)
+
+
+def _leaderboard_scheduler_sql_for_log(stmt) -> str:
+    try:
+        return str(stmt.compile(compile_kwargs={"literal_binds": True}))
+    except Exception:
+        return str(stmt)
+
+
+async def _execute_leaderboard_scheduler_query(session, step: str, stmt, **context):
+    sql = _leaderboard_scheduler_sql_for_log(stmt)
+    logger.info(
+        "Leaderboard snapshot DB query start: step=%s stat_date=%s job=%s settings_hash=%s started_at=%s sql=%s",
+        step,
+        context.get("stat_date") or context.get("stat_dates") or "",
+        context.get("job_key") or "",
+        context.get("settings_hash") or "",
+        datetime.now().isoformat(),
+        sql,
+    )
+    start = perf_counter()
+    try:
+        result = await session.execute(stmt)
+    except Exception:
+        logger.exception(
+            "Leaderboard snapshot DB query failed: step=%s stat_date=%s job=%s settings_hash=%s elapsed_ms=%.2f",
+            step,
+            context.get("stat_date") or context.get("stat_dates") or "",
+            context.get("job_key") or "",
+            context.get("settings_hash") or "",
+            (perf_counter() - start) * 1000,
+        )
+        raise
+    logger.info(
+        "Leaderboard snapshot DB query done: step=%s stat_date=%s job=%s settings_hash=%s finished_at=%s elapsed_ms=%.2f",
+        step,
+        context.get("stat_date") or context.get("stat_dates") or "",
+        context.get("job_key") or "",
+        context.get("settings_hash") or "",
+        datetime.now().isoformat(),
+        (perf_counter() - start) * 1000,
+    )
+    return result
+
+
+async def _time_leaderboard_scheduler_db_operation(step: str, operation, *, sql: str = "", **context):
+    logger.info(
+        "Leaderboard snapshot DB query start: step=%s stat_date=%s job=%s settings_hash=%s started_at=%s sql=%s",
+        step,
+        context.get("stat_date") or context.get("stat_dates") or "",
+        context.get("job_key") or "",
+        context.get("settings_hash") or "",
+        datetime.now().isoformat(),
+        sql,
+    )
+    start = perf_counter()
+    try:
+        result = await operation()
+    except Exception:
+        logger.exception(
+            "Leaderboard snapshot DB query failed: step=%s stat_date=%s job=%s settings_hash=%s elapsed_ms=%.2f",
+            step,
+            context.get("stat_date") or context.get("stat_dates") or "",
+            context.get("job_key") or "",
+            context.get("settings_hash") or "",
+            (perf_counter() - start) * 1000,
+        )
+        raise
+    logger.info(
+        "Leaderboard snapshot DB query done: step=%s stat_date=%s job=%s settings_hash=%s finished_at=%s elapsed_ms=%.2f",
+        step,
+        context.get("stat_date") or context.get("stat_dates") or "",
+        context.get("job_key") or "",
+        context.get("settings_hash") or "",
+        datetime.now().isoformat(),
+        (perf_counter() - start) * 1000,
+    )
+    return result
+
+
+def _recent_leaderboard_stat_dates(days: int) -> list[str]:
+    from datetime import datetime, timedelta
+
+    from deerflow.persistence.types import BEIJING_TZ
+
+    today = datetime.now(BEIJING_TZ).date()
+    start = today - timedelta(days=max(1, days) - 1)
+    return [(start + timedelta(days=i)).isoformat() for i in range((today - start).days + 1)]
+
+
+async def _leaderboard_snapshot_due_dates(
+    session_factory,
+    *,
+    settings_hash: str,
+    limit: int,
+    running_ttl_seconds: int,
+) -> list[str]:
+    from datetime import UTC, datetime, timedelta
+
+    from sqlalchemy import and_, or_, select
+
+    from deerflow.persistence.admin_stats.model import AdminLeaderboardDailyStatRow
+
+    cutoff = datetime.now(UTC) - timedelta(seconds=running_ttl_seconds)
+    async with session_factory() as session:
+        stmt = (
+            select(AdminLeaderboardDailyStatRow.stat_date)
+            .where(
+                AdminLeaderboardDailyStatRow.settings_hash == settings_hash,
+                or_(
+                    AdminLeaderboardDailyStatRow.status.in_(["queued", "error"]),
+                    and_(
+                        AdminLeaderboardDailyStatRow.status == "running",
+                        AdminLeaderboardDailyStatRow.updated_at < cutoff,
+                    ),
+                ),
+            )
+            .order_by(AdminLeaderboardDailyStatRow.stat_date.desc())
+            .limit(limit)
+        )
+        rows = (
+            await _execute_leaderboard_scheduler_query(
+                session,
+                "scheduler.due_dates",
+                stmt,
+                settings_hash=settings_hash,
+            )
+        ).all()
+    return [str(row[0]) for row in rows]
+
+
+async def _claim_leaderboard_snapshot(
+    session_factory,
+    *,
+    stat_date: str,
+    settings_hash: str,
+    running_ttl_seconds: int,
+) -> bool:
+    from datetime import UTC, datetime
+
+    from app.gateway.routers.admin_users import _snapshot_is_fresh
+    from deerflow.persistence.admin_stats.model import AdminLeaderboardDailyStatRow
+    from deerflow.persistence.types import BEIJING_TZ
+
+    async with session_factory() as session:
+        row = await _time_leaderboard_scheduler_db_operation(
+            "scheduler.claim.get",
+            lambda: session.get(
+                AdminLeaderboardDailyStatRow,
+                {"stat_date": stat_date, "settings_hash": settings_hash},
+            ),
+            sql="SELECT admin_leaderboard_daily_stats by primary key",
+            stat_date=stat_date,
+            settings_hash=settings_hash,
+        )
+        if row is None or _snapshot_is_fresh(row, stat_date):
+            return False
+        if row.status == "running":
+            updated_at = row.updated_at
+            if isinstance(updated_at, datetime):
+                if updated_at.tzinfo is None:
+                    updated_at = updated_at.replace(tzinfo=BEIJING_TZ)
+                age = (datetime.now(BEIJING_TZ) - updated_at.astimezone(BEIJING_TZ)).total_seconds()
+                if age < running_ttl_seconds:
+                    return False
+        row.status = "running"
+        row.message = "?????????????"
+        row.error = None
+        row.updated_at = datetime.now(UTC)
+        await _time_leaderboard_scheduler_db_operation(
+            "scheduler.claim.commit",
+            lambda: session.commit(),
+            sql="COMMIT",
+            stat_date=stat_date,
+            settings_hash=settings_hash,
+        )
+    return True
+
+
+async def _maybe_finalize_previous_leaderboard_day(session_factory, settings_hash: str) -> bool:
+    """Once per day after ``_LEADERBOARD_FINALIZE_HOUR``, force-recompute yesterday.
+
+    A ``ready`` past-day snapshot is otherwise treated as fresh forever and never
+    rebuilt. But the previous day's snapshot is usually generated before midnight,
+    so it can miss runs/tool-calls that only land in the DB around the day
+    boundary. This re-queues yesterday (``force=True``) a single time; the rebuild
+    stamps ``generated_at`` to today, which stops it from re-firing for the rest of
+    the day.
+    """
+    from datetime import timedelta
+
+    from app.gateway.routers.admin_users import (
+        _as_beijing,
+        _queue_daily_leaderboard_snapshots,
+    )
+    from deerflow.persistence.admin_stats.finalize import should_finalize_previous_day
+    from deerflow.persistence.admin_stats.model import AdminLeaderboardDailyStatRow
+    from deerflow.persistence.types import BEIJING_TZ
+
+    now = datetime.now(BEIJING_TZ)
+    yesterday = (now.date() - timedelta(days=1)).isoformat()
+    async with session_factory() as session:
+        row = await session.get(
+            AdminLeaderboardDailyStatRow,
+            {"stat_date": yesterday, "settings_hash": settings_hash},
+        )
+    generated_at = _as_beijing(row.generated_at) if (row is not None and row.generated_at is not None) else None
+    if not should_finalize_previous_day(
+        status=(row.status if row is not None else None),
+        generated_at=generated_at,
+        now=now,
+        finalize_hour=_LEADERBOARD_FINALIZE_HOUR,
+    ):
+        return False
+    queued = await _queue_daily_leaderboard_snapshots(
+        session_factory,
+        [yesterday],
+        settings_hash,
+        force=True,
+        message="???????????????????????????",
+    )
+    if queued:
+        logger.info(
+            "Leaderboard snapshot finalize: queued previous day %s for full recompute (hash=%s)",
+            yesterday,
+            settings_hash[:10],
+        )
+        return True
+    return False
+
+
+async def _run_leaderboard_snapshot_tick() -> tuple[int, int]:
+    from app.gateway.routers.admin_users import (
+        _daily_job_key,
+        _queue_daily_leaderboard_snapshots,
+        _refresh_daily_leaderboard_snapshot,
+        _resolve_leaderboard_snapshot_scope,
+    )
+    from deerflow.config.system_settings import load_system_settings
+    from deerflow.persistence.engine import get_session_factory
+
+    session_factory = get_session_factory()
+    if session_factory is None:
+        return 0, 0
+
+    settings = load_system_settings().leaderboard
+    excluded, include_scheduled, include_admins, include_failed, settings_hash = _resolve_leaderboard_snapshot_scope(settings)
+    stat_dates = _recent_leaderboard_stat_dates(_LEADERBOARD_SNAPSHOT_PREWARM_DAYS)
+    queued_dates = await _queue_daily_leaderboard_snapshots(
+        session_factory,
+        stat_dates,
+        settings_hash,
+        message="?????????????/?????",
+    )
+
+    # ??????? 1 ???????"???"????????????????????
+    try:
+        await _maybe_finalize_previous_leaderboard_day(session_factory, settings_hash)
+    except Exception:
+        logger.exception("Leaderboard snapshot finalize check failed (non-fatal)")
+
+    due_dates = await _leaderboard_snapshot_due_dates(
+        session_factory,
+        settings_hash=settings_hash,
+        limit=_LEADERBOARD_SNAPSHOT_BATCH_SIZE,
+        running_ttl_seconds=_LEADERBOARD_SNAPSHOT_RUNNING_TTL_SECONDS,
+    )
+    processed = 0
+    for stat_date in due_dates:
+        claimed = await _claim_leaderboard_snapshot(
+            session_factory,
+            stat_date=stat_date,
+            settings_hash=settings_hash,
+            running_ttl_seconds=_LEADERBOARD_SNAPSHOT_RUNNING_TTL_SECONDS,
+        )
+        if not claimed:
+            continue
+        logger.info("Leaderboard snapshot scheduler: building date=%s hash=%s", stat_date, settings_hash[:10])
+        await _refresh_daily_leaderboard_snapshot(
+            _daily_job_key(stat_date, settings_hash),
+            session_factory=session_factory,
+            settings=settings,
+            stat_date=stat_date,
+            settings_hash=settings_hash,
+            excluded=excluded,
+            include_scheduled=include_scheduled,
+            include_admins=include_admins,
+            include_failed=include_failed,
+        )
+        processed += 1
+        if _LEADERBOARD_SNAPSHOT_SLEEP_SECONDS > 0 and processed < len(due_dates):
+            await asyncio.sleep(_LEADERBOARD_SNAPSHOT_SLEEP_SECONDS)
+    return processed, len(queued_dates)
+
+
+async def _leaderboard_snapshot_scheduler_loop() -> None:
+    """Background task inside the existing gateway process for daily snapshots."""
+    if not _LEADERBOARD_SNAPSHOT_BACKGROUND_ENABLED:
+        logger.info("Leaderboard snapshot scheduler disabled by ADMIN_LEADERBOARD_BACKGROUND_ENABLED")
+        return
+
+    from deerflow.config.runtime_paths import runtime_home
+
+    lock_handle = _acquire_singleton_lock(runtime_home() / ".leaderboard_snapshot_scheduler.lock")
+    if lock_handle is None:
+        logger.info("Leaderboard snapshot scheduler: another gateway worker holds the lock; this worker skips")
+        return
+
+    logger.info(
+        "Leaderboard snapshot scheduler started: prewarm_days=%d batch_size=%d interval=%.1fs",
+        _LEADERBOARD_SNAPSHOT_PREWARM_DAYS,
+        _LEADERBOARD_SNAPSHOT_BATCH_SIZE,
+        _LEADERBOARD_SNAPSHOT_INTERVAL_SECONDS,
+    )
+    try:
+        if _LEADERBOARD_SNAPSHOT_STARTUP_DELAY_SECONDS > 0:
+            await asyncio.sleep(_LEADERBOARD_SNAPSHOT_STARTUP_DELAY_SECONDS)
+        while True:
+            try:
+                processed, queued = await _run_leaderboard_snapshot_tick()
+                if processed or queued:
+                    logger.info(
+                        "Leaderboard snapshot scheduler tick: queued=%d processed=%d",
+                        queued,
+                        processed,
+                    )
+            except Exception:
+                logger.exception("Leaderboard snapshot scheduler tick failed (non-fatal)")
+            await asyncio.sleep(_LEADERBOARD_SNAPSHOT_INTERVAL_SECONDS)
+    finally:
+        _release_singleton_lock(lock_handle)
+
+
+async def _concurrency_sampler_loop(app: FastAPI) -> None:
+    """Periodically record the in-flight run count for the admin pressure chart.
+
+    Runs in the gateway process, singleton-locked so only one worker samples.
+    Each tick records ``len(run_manager.list_active())`` and, roughly hourly,
+    purges samples older than the retention window. Best-effort throughout — a
+    failed tick is logged and the loop continues.
+    """
+    if not _CONCURRENCY_SAMPLE_ENABLED:
+        logger.info("Concurrency sampler disabled by CONCURRENCY_SAMPLE_ENABLED")
+        return
+
+    store = getattr(app.state, "concurrency_sample_store", None)
+    run_mgr = getattr(app.state, "run_manager", None)
+    concurrency_gate = getattr(app.state, "concurrency_gate", None)
+    if store is None or run_mgr is None:
+        logger.info("Concurrency sampler not started (no store / run manager — memory backend?)")
+        return
+
+    from deerflow.config.runtime_paths import runtime_home
+
+    lock_handle = _acquire_singleton_lock(runtime_home() / ".concurrency_sampler.lock")
+    if lock_handle is None:
+        logger.info("Concurrency sampler: another gateway worker holds the lock; this worker skips")
+        return
+
+    interval = max(1.0, _CONCURRENCY_SAMPLE_INTERVAL_SECONDS)
+    purge_every_ticks = max(1, int(3600 / interval))  # ~once per hour
+    logger.info("Concurrency sampler started: interval=%.1fs retention=%.1fh", interval, _CONCURRENCY_SAMPLE_RETENTION_HOURS)
+    ticks = 0
+    try:
+        while True:
+            try:
+                # Runtime toggle (admin can turn sampling off without a restart).
+                # mtime-cached read, so this is cheap every tick.
+                from deerflow.config.system_settings import get_concurrency_monitor_settings
+
+                if get_concurrency_monitor_settings().enabled:
+                    active_count = None
+                    if concurrency_gate is not None:
+                        active_count = await concurrency_gate.count_active()
+                    if active_count is None:
+                        active = await run_mgr.list_active()
+                        active_count = len(active)
+                    await store.record(active_count)
+                    ticks += 1
+                    if ticks % purge_every_ticks == 0:
+                        from datetime import UTC, datetime, timedelta
+
+                        cutoff = datetime.now(UTC) - timedelta(hours=_CONCURRENCY_SAMPLE_RETENTION_HOURS)
+                        await store.purge_older_than(cutoff)
+            except Exception:
+                logger.exception("Concurrency sampler tick failed (non-fatal)")
+            await asyncio.sleep(interval)
+    finally:
+        _release_singleton_lock(lock_handle)
+
+
+def _warn_if_multi_worker() -> None:
+    """?? uvicorn / gunicorn ? worker ???? ERROR ???
+
+    ? agent / canvas / AI ??????? Gateway ????? ``StreamBridge``
+    ?? run_id ? SSE ?????? map?? worker ????? worker A ??
+    ``/runs/stream``???? worker B ? join/resume??? in-process map ??
+    ???? 404 / ???
+
+    ????? startup ? ERROR ????**???? fail-fast** ?? ??
+    ???????????????? sticky session??????????
+    """
+    # uvicorn / gunicorn ??????? worker ?????????????
+    # ?????? sibling ??????????? WEB_CONCURRENCY?gunicorn
+    # / uvicorn ????
+    web_concurrency = os.environ.get("WEB_CONCURRENCY", "").strip()
+    if web_concurrency and web_concurrency.isdigit() and int(web_concurrency) > 1:
+        logger.error(
+            "? ??? WEB_CONCURRENCY=%s?? worker ????AI ?? / ? agent "
+            "??????????????? worker ????????? / "
+            "SSE ?? / resume 400??????????? WEB_CONCURRENCY=1 "
+            "????????? sticky session??? AIWritingPage.tsx ????",
+            web_concurrency,
+        )
+        return
+
+    # ???? /proc ??? uvicorn ?????? Linux ??????
+    if not sys.platform.startswith("linux"):
+        return
+    try:
+        from pathlib import Path
+
+        peers = 0
+        for entry in Path("/proc").iterdir():
+            if not entry.name.isdigit():
+                continue
+            try:
+                cmdline = (entry / "cmdline").read_bytes().replace(b"\x00", b" ").decode("utf-8", "ignore")
+            except OSError:
+                continue
+            if "uvicorn" in cmdline and "app.gateway.app" in cmdline:
+                peers += 1
+        if peers > 1:
+            logger.error(
+                "? ??? %d ??? uvicorn ???? worker ????? AI ?? "
+                "????? / SSE ??????????? worker ??????? "
+                "?? sticky session?",
+                peers,
+            )
+    except Exception:
+        # ????????? ?? ??????
+        pass
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI) -> AsyncGenerator[None, None]:
+    """Application lifespan handler."""
+
+    # ?????? worker ?? ?? ?? AI ?? / ? agent ???????
+    # ??????? worker ???????? ERROR ?????
+    _warn_if_multi_worker()
+
+    # Load config and check necessary environment variables at startup
+    try:
+        app.state.config = get_app_config()
+        apply_logging_level(app.state.config.log_level)
+        logger.info("Configuration loaded successfully")
+    except Exception as e:
+        error_msg = f"Failed to load configuration during gateway startup: {e}"
+        logger.exception(error_msg)
+        raise RuntimeError(error_msg) from e
+    config = get_gateway_config()
+    logger.info(f"Starting API Gateway on {config.host}:{config.port}")
+
+    # Initialize LangGraph runtime components (StreamBridge, RunManager, checkpointer, store)
+    async with langgraph_runtime(app):
+        logger.info("LangGraph runtime initialised")
+
+        # Run Alembic migrations before anything else touches the schema.
+        await _run_db_migrations()
+
+        # ?? recall_budget cache?? DB ????,? get_effective_memory_config ?????
+        try:
+            from deerflow.config.memory_config import load_recall_budget_cache
+
+            _mc_store = getattr(app.state, "memory_config_store", None)
+            if _mc_store is not None:
+                _recall_budgets = await _mc_store.load_all()
+                load_recall_budget_cache(_recall_budgets)
+                logger.info("Memory config cache warmed: %d user override(s)", len(_recall_budgets))
+        except Exception:
+            logger.exception("Memory config cache warm-up failed (non-fatal)")
+
+        # Ensure admin user exists (auto-create on first boot)
+        # Must run AFTER langgraph_runtime so app.state.store is available for thread migration
+        await _ensure_admin_user(app)
+
+        # Migrate "ownerless" on-disk skills into the new SkillStore table ?
+        # makes the multitenancy switch transparent for existing deployments.
+        try:
+            await _sync_legacy_skills(app)
+        except Exception:
+            logger.exception("Skill ownership migration failed (non-fatal)")
+
+        # ????? agent(intent / recommender / coordinator)???????
+        # ??? _sync_legacy_agents **??**??????????????????
+        # ??? upsert ? agents ?,?????? /api/agents ?????,
+        # ????????????????.deer-flow/ ? gitignore,??? clone
+        # ????????,seeder ????? _roundtable_seed_assets/ ??????
+        try:
+            from app.gateway.routers._roundtable_seed import (
+                ensure_roundtable_functional_agents,
+            )
+
+            ensure_roundtable_functional_agents()
+        except Exception:
+            logger.exception(
+                "Roundtable functional agent seeding failed at startup (non-fatal)"
+            )
+
+        # AI ????? agent(researcher / outliner / writer / editor)???????
+        # ???:??? _sync_legacy_agents ??,?????????????
+        # upsert ? agents ?,admin ?????????????????? agent?
+        try:
+            from app.gateway.routers._ai_writing_seed import (
+                ensure_ai_writing_functional_agents,
+            )
+
+            ensure_ai_writing_functional_agents()
+        except Exception:
+            logger.exception(
+                "AI-writing functional agent seeding failed at startup (non-fatal)"
+            )
+
+        # 内置「模板数据转换器」agent(JSON → HTML 模板填充)。同样必须在
+        # _sync_legacy_agents **之前**补建目录,后者把它 upsert 进 agents 表,
+        # 第一次启动 /api/agents 与定时任务下拉框里就有它。
+        try:
+            from app.gateway.routers._template_builder_seed import (
+                ensure_template_builder_agent,
+            )
+
+            ensure_template_builder_agent()
+        except Exception:
+            logger.exception(
+                "Template-builder agent seeding failed at startup (non-fatal)"
+            )
+
+        # 深度研究「收集员」内置 agent(对话式信息收集)。同样必须在
+        # _sync_legacy_agents 之前补建目录,由后者 upsert 进 agents 表,
+        # 深度研究页的收集对话与智能体管理页即可直接使用。
+        try:
+            from app.gateway.routers._deep_research_seed import (
+                ensure_deep_research_collector_agent,
+            )
+
+            ensure_deep_research_collector_agent()
+        except Exception:
+            logger.exception(
+                "Deep-research collector agent seeding failed at startup (non-fatal)"
+            )
+
+        # ?????????????????3Q/6BF/7BF/8BF??agent/chain ?????
+        try:
+            business_mapping_store = getattr(app.state, "business_mapping_store", None)
+            if business_mapping_store is not None:
+                inserted = await business_mapping_store.ensure_seed_defaults()
+                if inserted:
+                    logger.info("Seeded %d business mapping rows", inserted)
+        except Exception:
+            logger.exception("Business mapping seeding failed at startup (non-fatal)")
+
+        # ???? .deer-flow/agents/{id}/ ???????? agent ?? agents ??
+        # ??????????????? sync(????? + N ? agent ??? DB
+        # SELECT/UPDATE),??????? + ?????? TTL ??,????
+        # /api/agents ?????
+        try:
+            from app.gateway.routers.agents import (
+                _mark_legacy_sync_fresh,
+                _sync_legacy_agents,
+            )
+            from app.gateway.services import _mark_legacy_sync_fresh as _services_mark_sync_fresh
+
+            agent_store = getattr(app.state, "agent_store", None)
+            if agent_store is not None:
+                await _sync_legacy_agents(agent_store)
+                _mark_legacy_sync_fresh()
+                # ``services.py`` keeps an independent TTL gate (used by
+                # ``start_run`` ? ``_enforce_agent_access``). Without this
+                # ack, the first ``/runs/stream`` after startup would
+                # re-run the heavy sync even though we just did it.
+                _services_mark_sync_fresh()
+                logger.info("Legacy folder-backed agents synced at startup")
+                # 内置功能型 agent(roundtable-* / ai-writing-*)按用途默认打标签,
+                # 供管理员「内置智能体」筛选按用途归类。幂等、只追加、失败不阻塞启动。
+                try:
+                    from app.gateway.routers._builtin_agent_tag_seed import (
+                        ensure_builtin_agent_purpose_tags,
+                    )
+
+                    tag_store = getattr(app.state, "tag_store", None)
+                    if tag_store is not None:
+                        await ensure_builtin_agent_purpose_tags(agent_store, tag_store)
+                except Exception:
+                    logger.exception("Built-in agent purpose-tag seeding failed at startup (non-fatal)")
+        except Exception:
+            logger.exception("Legacy agent sync failed at startup (non-fatal)")
+
+        # ?? skill curator ???:??????,?????????????
+        # (per-user interval_hours ??????????????)?
+        try:
+            app.state.curator_scheduler_task = asyncio.create_task(_curator_scheduler_loop())
+        except Exception:
+            logger.exception("Skill curator scheduler failed to start (non-fatal)")
+
+        # ?? AI ????????:? config.yaml ? ai_writing.cleanup ???,
+        # ?????? 3 ????,? retention_days(?? 7 ?)?????,??
+        # ?? MySQL ??? + SQLite checkpointer ?? thread ???? worker
+        # ?????? leader,?????? worker ??????
+        try:
+            from app.gateway.ai_writing_cleanup import cleanup_scheduler_loop
+
+            app.state.ai_writing_cleanup_task = asyncio.create_task(cleanup_scheduler_loop(app))
+        except Exception:
+            logger.exception("AI writing cleanup scheduler failed to start (non-fatal)")
+
+        try:
+            app.state.leaderboard_snapshot_scheduler_task = asyncio.create_task(_leaderboard_snapshot_scheduler_loop())
+        except Exception:
+            logger.exception("Leaderboard snapshot scheduler failed to start (non-fatal)")
+
+        # Concurrency (system-pressure) sampler — lightweight, singleton-locked,
+        # runtime-toggleable via system_settings.concurrency_monitor.enabled.
+        try:
+            app.state.concurrency_sampler_task = asyncio.create_task(_concurrency_sampler_loop(app))
+        except Exception:
+            logger.exception("Concurrency sampler failed to start (non-fatal)")
+
+        # Start IM channel service if any channels are configured
+        try:
+            from app.channels.service import start_channel_service
+
+            channel_service = await start_channel_service(app.state.config)
+            logger.info("Channel service started: %s", channel_service.get_status())
+        except Exception:
+            logger.exception("No IM channels configured or channel service failed to start")
+
+        scheduled_tasks_enabled = os.environ.get("SCHEDULED_TASKS_ENABLED", "true").strip().lower()
+        if scheduled_tasks_enabled in {"0", "false", "no", "off"}:
+            logger.info("Scheduled task service disabled by SCHEDULED_TASKS_ENABLED=%s", scheduled_tasks_enabled)
+        else:
+            try:
+                scheduled_task_store = getattr(app.state, "scheduled_task_store", None)
+                if scheduled_task_store is not None:
+                    scheduler_service = ScheduledTaskService(app, scheduled_task_store)
+                    app.state.scheduled_task_service = scheduler_service
+                    set_scheduled_task_service(scheduler_service)
+                    await scheduler_service.start()
+                    logger.info("Scheduled task service started")
+            except Exception:
+                logger.exception("Scheduled task service failed to start")
+
+        yield
+
+        # ?? curator ???
+        curator_task = getattr(app.state, "curator_scheduler_task", None)
+        if curator_task is not None:
+            curator_task.cancel()
+            try:
+                await curator_task
+            except asyncio.CancelledError:
+                pass
+            except Exception:
+                logger.exception("Curator scheduler shutdown error")
+
+        # ?? AI ???????
+        cleanup_task = getattr(app.state, "ai_writing_cleanup_task", None)
+        if cleanup_task is not None:
+            cleanup_task.cancel()
+            try:
+                await cleanup_task
+            except asyncio.CancelledError:
+                pass
+            except Exception:
+                logger.exception("AI writing cleanup scheduler shutdown error")
+
+        leaderboard_snapshot_task = getattr(app.state, "leaderboard_snapshot_scheduler_task", None)
+        if leaderboard_snapshot_task is not None:
+            leaderboard_snapshot_task.cancel()
+            try:
+                await leaderboard_snapshot_task
+            except asyncio.CancelledError:
+                pass
+            except Exception:
+                logger.exception("Leaderboard snapshot scheduler shutdown error")
+
+        concurrency_sampler_task = getattr(app.state, "concurrency_sampler_task", None)
+        if concurrency_sampler_task is not None:
+            concurrency_sampler_task.cancel()
+            try:
+                await concurrency_sampler_task
+            except asyncio.CancelledError:
+                pass
+            except Exception:
+                logger.exception("Concurrency sampler shutdown error")
+
+        # ?? 7 ????AI ?? graph ???? checkpointer / in-flight task?
+        # thread ??? LangGraph Server ? checkpointer ???lifespan ???
+        # ``langgraph_runtime()`` ? contextmanager ???????????????
+
+        try:
+            scheduler_service = getattr(app.state, "scheduled_task_service", None)
+            if scheduler_service is not None:
+                await scheduler_service.stop()
+        except Exception:
+            logger.exception("Failed to stop scheduled task service")
+        finally:
+            set_scheduled_task_service(None)
+
+        # Stop channel service on shutdown (bounded to prevent worker hang)
+        try:
+            from app.channels.service import stop_channel_service
+
+            await asyncio.wait_for(
+                stop_channel_service(),
+                timeout=_SHUTDOWN_HOOK_TIMEOUT_SECONDS,
+            )
+        except TimeoutError:
+            logger.warning(
+                "Channel service shutdown exceeded %.1fs; proceeding with worker exit.",
+                _SHUTDOWN_HOOK_TIMEOUT_SECONDS,
+            )
+        except Exception:
+            logger.exception("Failed to stop channel service")
+
+    logger.info("Shutting down API Gateway")
+
+
+def create_app() -> FastAPI:
+    """Create and configure the FastAPI application.
+
+    Returns:
+        Configured FastAPI application instance.
+    """
+    config = get_gateway_config()
+    docs_kwargs = {"docs_url": "/docs", "redoc_url": "/redoc", "openapi_url": "/openapi.json"} if config.enable_docs else {"docs_url": None, "redoc_url": None, "openapi_url": None}
+
+    app = FastAPI(
+        title="DeerFlow API Gateway",
+        description="""
+## DeerFlow API Gateway
+
+API Gateway for DeerFlow - A LangGraph-based AI agent backend with sandbox execution capabilities.
+
+### Features
+
+- **Models Management**: Query and retrieve available AI models
+- **MCP Configuration**: Manage Model Context Protocol (MCP) server configurations
+- **Memory Management**: Access and manage global memory data for personalized conversations
+- **Skills Management**: Query and manage skills and their enabled status
+- **Artifacts**: Access thread artifacts and generated files
+- **Health Monitoring**: System health check endpoints
+
+### Architecture
+
+LangGraph requests are handled by nginx reverse proxy.
+This gateway provides custom endpoints for models, MCP configuration, skills, and artifacts.
+        """,
+        version="0.1.0",
+        lifespan=lifespan,
+        **docs_kwargs,
+        openapi_tags=[
+            {
+                "name": "models",
+                "description": "Operations for querying available AI models and their configurations",
+            },
+            {
+                "name": "mcp",
+                "description": "Manage Model Context Protocol (MCP) server configurations",
+            },
+            {
+                "name": "memory",
+                "description": "Access and manage global memory data for personalized conversations",
+            },
+            {
+                "name": "skills",
+                "description": "Manage skills and their configurations",
+            },
+            {
+                "name": "artifacts",
+                "description": "Access and download thread artifacts and generated files",
+            },
+            {
+                "name": "uploads",
+                "description": "Upload and manage user files for threads",
+            },
+            {
+                "name": "threads",
+                "description": "Manage DeerFlow thread-local filesystem data",
+            },
+            {
+                "name": "agents",
+                "description": "Create and manage custom agents with per-agent config and prompts",
+            },
+            {
+                "name": "suggestions",
+                "description": "Generate follow-up question suggestions for conversations",
+            },
+            {
+                "name": "channels",
+                "description": "Manage IM channel integrations (Feishu, Slack, Telegram)",
+            },
+            {
+                "name": "assistants-compat",
+                "description": "LangGraph Platform-compatible assistants API (stub)",
+            },
+            {
+                "name": "runs",
+                "description": "LangGraph Platform-compatible runs lifecycle (create, stream, cancel)",
+            },
+            {
+                "name": "scheduled-tasks",
+                "description": "Create, manage, and execute user scheduled agent tasks",
+            },
+            {
+                "name": "health",
+                "description": "Health check and system status endpoints",
+            },
+        ],
+    )
+
+    # Auth: reject unauthenticated requests to non-public paths (fail-closed safety net)
+    app.add_middleware(AuthMiddleware)
+
+    # CSRF: Double Submit Cookie pattern for state-changing requests
+    app.add_middleware(CSRFMiddleware)
+
+    # CORS
+    if is_auth_disabled():
+        logger.warning(
+            "DEER_FLOW_AUTH_DISABLED is set ? API authentication is bypassed. "
+            "Do not use this in production."
+        )
+
+    def _cors_allow_any_origin() -> bool:
+        """Mirror any browser Origin (works with credentials); cannot use literal '*'."""
+        v = os.environ.get("GATEWAY_CORS_ALLOW_ALL", "").strip().lower()
+        return v in ("1", "true", "yes", "on") or is_auth_disabled()
+
+    # Keep standard local frontend origins available even when the startup
+    # script supplies an explicit production/development allow-list. Vite can
+    # choose 5173 or move to 5174 when that port is already occupied; excluding
+    # the former makes every browser preflight fail after a backend restart.
+    local_vite_origins = [
+        "http://localhost:5173",
+        "http://127.0.0.1:5173",
+        "http://localhost:5174",
+        "http://127.0.0.1:5174",
+        "http://localhost:3000",
+        "http://127.0.0.1:3000",
+    ]
+
+    # The legacy TaskCOP Vue application uses its own Vite development port.
+    taskcop_vite_origins = [
+        "http://localhost:8080",
+        "http://127.0.0.1:8080",
+    ]
+    local_dev_origins = [*local_vite_origins, *taskcop_vite_origins]
+
+    cors_origins_env = os.environ.get("GATEWAY_CORS_ORIGINS", "").strip()
+    cors_origins: list[str] = [o.strip() for o in cors_origins_env.split(",") if o.strip()] if cors_origins_env else []
+    if cors_origins:
+        cors_origins = list(dict.fromkeys([*cors_origins, *local_dev_origins]))
+
+    # Headers the LangGraph SDK / browser clients need to read from cross-origin
+    # responses. Without exposing them, the SDK's `response.headers.get(...)` returns
+    # null on cross-origin streaming responses, breaking run-id extraction and reconnect.
+    cors_expose_headers = [
+        "Content-Disposition",
+        "Content-Location",
+        "Location",
+        "X-Run-Id",
+        "X-Thread-Id",
+    ]
+
+    # Public pages (e.g. http://47.x.x.x:7010) calling this Gateway on localhost
+    # trigger Chrome/Edge Private Network Access preflights. Without
+    # allow_private_network=True the browser reports a generic CORS failure even
+    # when Access-Control-Allow-Origin is correct.
+    cors_kwargs = {
+        "allow_credentials": True,
+        "allow_methods": ["*"],
+        "allow_headers": ["*"],
+        "expose_headers": cors_expose_headers,
+        "allow_private_network": True,
+    }
+
+    if _cors_allow_any_origin():
+        logger.warning(
+            "Permissive CORS enabled ? any Origin is echoed back (GATEWAY_CORS_ALLOW_ALL or "
+            "DEER_FLOW_AUTH_DISABLED). Do not expose publicly."
+        )
+        app.add_middleware(
+            CORSMiddleware,
+            allow_origins=[],
+            # fullmatch() on the Origin header; browsers send scheme://host[:port]
+            allow_origin_regex=r".+",
+            **cors_kwargs,
+        )
+    elif cors_origins:
+        if "*" in cors_origins:
+            logger.error(
+                "GATEWAY_CORS_ORIGINS contains '*' with allow_credentials=True ? invalid. "
+                "Set GATEWAY_CORS_ALLOW_ALL=1 instead, or list explicit origins."
+            )
+            cors_origins = [o for o in cors_origins if o != "*"]
+        if cors_origins:
+            app.add_middleware(
+                CORSMiddleware,
+                allow_origins=cors_origins,
+                **cors_kwargs,
+            )
+    else:
+        # Safe-by-default fallback for local frontend development. Without any CORS
+        # middleware, browser preflight OPTIONS requests will hit routes directly and
+        # fail with 405, which blocks login and all JSON POST APIs.
+        dev_origins = local_dev_origins
+        logger.warning(
+            "No CORS configuration found; enabling local dev origins only: %s. "
+            "Set GATEWAY_CORS_ORIGINS or GATEWAY_CORS_ALLOW_ALL for deployment.",
+            ", ".join(dev_origins),
+        )
+        app.add_middleware(
+            CORSMiddleware,
+            allow_origins=dev_origins,
+            **cors_kwargs,
+        )
+
+    # Include routers
+    # Models API is mounted at /api/models
+    app.include_router(models.router)
+    # Administrator-managed config.yaml model settings and connection tests.
+    app.include_router(model_management.router)
+
+    # MCP API is mounted at /api/mcp
+    app.include_router(mcp.router)
+
+    # DeerFlow-native LLMWiki management/search BFF (conditionally backed by WeKnora).
+    app.include_router(llmwiki.router)
+
+    # Memory API is mounted at /api/memory
+    app.include_router(memory.router)
+
+    # Skills API is mounted at /api/skills
+    app.include_router(skills.router)
+
+    # Artifacts API is mounted at /api/threads/{thread_id}/artifacts
+    app.include_router(artifacts.router)
+
+    # Uploads API is mounted at /api/threads/{thread_id}/uploads
+    app.include_router(uploads.router)
+
+    # Optional Chrome-extension page capture API is mounted at /api/browser-context
+    app.include_router(browser_context.router)
+
+    # Thread cleanup API is mounted at /api/threads/{thread_id}
+    app.include_router(threads.router)
+
+    # Agents API is mounted at /api/agents
+    app.include_router(agents.router)
+
+    # Recommended Questions API is mounted at /api/recommended-questions
+    app.include_router(recommended_questions.router)
+
+    app.include_router(user_prompts.router)
+
+    # Per-user UI preferences at /api/user-preferences
+    app.include_router(user_preferences.router)
+
+    # Roundtable-planning drafts + recommendation history at /api/roundtable-drafts
+    app.include_router(roundtable_drafts.router)
+
+    # Task-scoped roundtable drafts (taskId 深链聊天记录，独立存、不分权) at /api/roundtable-task-drafts
+    app.include_router(roundtable_task_drafts.router)
+
+    # Task-scoped artifact submission status; stores references only, never artifact content.
+    app.include_router(roundtable_artifact_submissions.router)
+
+    # Step3 流程图结构化抽取 + 校验 (SSE) at /api/flow-extract
+    app.include_router(flow_extract.router)
+
+    # Roundtable-planning business chains at /api/roundtable-chains
+    app.include_router(roundtable_chains.router)
+
+    # 岗位协同独立会话 / 节点状态（不使用原圆桌总控或后台作业）
+    # Canonical position-workspace persistence API.
+    app.include_router(position_roundtable.router, prefix="/api")
+    # Compatibility route for intranet gateways whose allowlist only exposes
+    # the established `/api/multi-agent/*` namespace.
+    app.include_router(
+        position_roundtable.router,
+        prefix="/api/multi-agent",
+        include_in_schema=False,
+    )
+
+    # Global business ? agent / chain mapping at /api/business-mapping
+    app.include_router(business_mapping.router)
+
+    # Position-roundtable work-view role configuration at /api/position-roles.
+    app.include_router(position_roles.router)
+
+    # Light-app registry (?????) at /api/light-apps
+    app.include_router(light_apps.router)
+
     # Sidebar-menu overrides (菜单管理) at /api/menu-overrides
     app.include_router(menu_overrides.router)
 
@@ -1528,158 +1529,161 @@ This gateway provides custom endpoints for models, MCP configuration, skills, an
     app.include_router(compact_menu_overrides.router)
 
     # Sensitive-word map (敏感词管理 / 脱敏词) at /api/sensitive-words
-    app.include_router(sensitive_words.router)
-
-    # Skill URL/IP batch replace (技能地址管理) at /api/skill-addresses
-    app.include_router(skill_addresses.router)
-
-    # 大屏绘制智能体独立页面会话（按 taskId 存取，不分权）at /api/dashboard-sessions
-    app.include_router(dashboard_sessions.router)
-
-    # Task-workspace configurable jump buttons (按钮管理) at /api/task-buttons
-    app.include_router(task_buttons.router)
-
-    # Temporary compatibility bridge for the TaskCOP situation-overview frontend.
-    app.include_router(taskcop_tasks.router)
-
-    # Task-scoped situation-report details used by dashboard drawing / task gates.
-    app.include_router(task_reports.router)
-
-    # Standalone parallel multi-agent panel: fan-out streaming + saved tasks
-    app.include_router(parallel_agents.router)
-    app.include_router(parallel_agent_tasks.router)
-
-    # Roundtable-planning background jobs at /api/roundtable-jobs
-    app.include_router(roundtable_jobs.router)
-    app.include_router(roundtable_diagnostics.router)
-
-    # Deep Research (深度研究) API at /api/deep-research
-    app.include_router(deep_research.router)
-
-    # Suggestions API is mounted at /api/threads/{thread_id}/suggestions
-    app.include_router(suggestions.router)
-
-    # Writing rewrite API is mounted at /api/writing
-    app.include_router(writing.router)
-
-    # AI Writing multi-agent API is mounted at /api/ai-writing
-    app.include_router(ai_writing.router)
-
-    # Scheduled task API is mounted at /api/scheduled-tasks
-    app.include_router(scheduled_tasks.router)
-
-    # Public, unauthenticated read-only scheduled task API
-    # (/api/public/scheduled-tasks/*). Lets external systems query tasks and
-    # surface results without a DeerFlow login.
-    app.include_router(public_scheduled_tasks.router)
-
-    # Public read-only user leaderboard publish page (/api/public/leaderboard)
-    app.include_router(public_leaderboard.router)
-
-    # Public embed Q&A status polling (/api/public/embed/*)
-    app.include_router(public_embed.router)
-    app.include_router(public_config.router)
-    app.include_router(html_page_favorites.router)
-
-    # Product knowledge base API is mounted at /api/knowledge
-    app.include_router(knowledge.router)
-
-    # External knowledge ingest (publish md artifacts) at /api/knowledge-ingest
-    app.include_router(knowledge_ingest.router)
-
-    # System settings API is mounted at /api/system-settings
-    app.include_router(system_settings.router)
-
-    # Admin LLM metrics audit + export
-    app.include_router(llm_metrics.router)
-
-    # Admin window into other users' threads / messages / memory
-    app.include_router(admin_users.router)
-    app.include_router(admin_users.self_router)
-
-    app.include_router(admin_active_runs.router)
-
-    # Admin tool / skill call metrics audit + export
-    app.include_router(tool_metrics.router)
-
-    # Channels API is mounted at /api/channels
-    app.include_router(channels.router)
-
-    # Admin checkpoint migration API: SQLite legacy checkpoints -> Postgres.
-    app.include_router(checkpoint_migration.router)
-
-    # Assistants compatibility API (LangGraph Platform stub)
-    app.include_router(assistants_compat.router)
-
-    # Auth API is mounted at /api/v1/auth
-    app.include_router(auth.router)
-
-    # Feedback API is mounted at /api/threads/{thread_id}/runs/{run_id}/feedback
-    app.include_router(feedback.router)
-
-    # Notifications API is mounted at /api/notifications
-    app.include_router(notifications.router)
-
-    # Tags API is mounted at /api/tags
-    app.include_router(tags.router)
-
-    # 岗位 (position) management API is mounted at /api/positions
-    app.include_router(positions.router)
-
-    # Thread Runs API (LangGraph Platform-compatible runs lifecycle)
-    app.include_router(thread_runs.router)
-
-    app.include_router(thread_shares.router)
-    # Public, unauthenticated read-only share snapshots (/api/public/shares/*).
-    app.include_router(thread_shares.public_router)
-
-    # ?????????import ???? / view ????????
-    app.include_router(roundtable_draft_shares.router)
-    # Public, unauthenticated read-only roundtable result (/api/public/roundtable-shares/*).
-    app.include_router(roundtable_draft_shares.public_router)
-
-    # Stateless Runs API (stream/wait without a pre-existing thread)
-    app.include_router(runs.router)
-
-    # 开放问答接口 /api/open/chat —— 无需登录,指定模型 + 智能体做问答,可关闭思考模式
-    app.include_router(open_chat.router)
-
-    # 开放接口 /api/open/3qfx/ask —— 无需登录,按 taskId 直接发起 3qfx 第一次问答(后台运行,落成真实会话)
-    app.include_router(open_3qfx.router)
-
-    # --- ?????????? /page/workspace/roundtable/planning?---
-    # ??? app/gateway/routers/{multi_agent,intent,recommend}.py?
-    # ??? frontend-web/docs/multi-agent-backend-dev.md
-    app.include_router(multi_agent.router)   # Step 2?init(SSE) + run/stream(leader|special)
-    app.include_router(intent.router)        # Step 1?/api/intent/init + /stream
-    app.include_router(recommend.router)     # Step 1?2?/api/recommend/stream?? init?
-    # Step 3 ????/?????? artifacts.router?/api/threads/{id}/artifacts?
-
-    # Health probes are always unauthenticated (see
-    # auth_middleware.PUBLIC_HEALTH_PATHS). Support the bare Gateway path,
-    # API base URL, and LangGraph proxy base URL for deployment monitors.
-    @app.api_route("/health", methods=["GET", "HEAD"], tags=["health"])
-    @app.api_route("/api/health", methods=["GET", "HEAD"], tags=["health"])
-    @app.api_route("/api/langgraph/health", methods=["GET", "HEAD"], tags=["health"])
-    async def health_check() -> dict:
-        """Health check endpoint.
-
-        Returns:
-            Service health status information.
-        """
-        return {"status": "healthy", "service": "deer-flow-gateway"}
-
-    # WeKnora's embedded frontend uses root-relative routes such as /platform,
-    # /assets, /locales and /api/v1/*. Register that proxy last so it stays a
-    # true fallback and cannot intercept DeerFlow APIs.
-    # Register the public-prefix routes before the unprefixed static fallback.
-    # Otherwise proxy_router's /{proxied_path:path} route consumes every
-    # /deerflow/* request before the prefixed platform/assets routes can match.
-    app.include_router(llmwiki.proxy_router, prefix="/deerflow")
-    app.include_router(llmwiki.proxy_router)
-
-    return app
-
-
-# Create app instance for uvicorn
-app = create_app()
+    app.include_router(sensitive_words.router)
+
+    # Skill URL/IP batch replace (技能地址管理) at /api/skill-addresses
+    app.include_router(skill_addresses.router)
+
+    # 大屏绘制智能体独立页面会话（按 taskId 存取，不分权）at /api/dashboard-sessions
+    app.include_router(dashboard_sessions.router)
+
+    # Task-workspace configurable jump buttons (按钮管理) at /api/task-buttons
+    app.include_router(task_buttons.router)
+
+    # Deep-research report structures (工作流配置 / 报告结构) at /api/report-structures
+    app.include_router(report_structures.router)
+
+    # Temporary compatibility bridge for the TaskCOP situation-overview frontend.
+    app.include_router(taskcop_tasks.router)
+
+    # Task-scoped situation-report details used by dashboard drawing / task gates.
+    app.include_router(task_reports.router)
+
+    # Standalone parallel multi-agent panel: fan-out streaming + saved tasks
+    app.include_router(parallel_agents.router)
+    app.include_router(parallel_agent_tasks.router)
+
+    # Roundtable-planning background jobs at /api/roundtable-jobs
+    app.include_router(roundtable_jobs.router)
+    app.include_router(roundtable_diagnostics.router)
+
+    # Deep Research (深度研究) API at /api/deep-research
+    app.include_router(deep_research.router)
+
+    # Suggestions API is mounted at /api/threads/{thread_id}/suggestions
+    app.include_router(suggestions.router)
+
+    # Writing rewrite API is mounted at /api/writing
+    app.include_router(writing.router)
+
+    # AI Writing multi-agent API is mounted at /api/ai-writing
+    app.include_router(ai_writing.router)
+
+    # Scheduled task API is mounted at /api/scheduled-tasks
+    app.include_router(scheduled_tasks.router)
+
+    # Public, unauthenticated read-only scheduled task API
+    # (/api/public/scheduled-tasks/*). Lets external systems query tasks and
+    # surface results without a DeerFlow login.
+    app.include_router(public_scheduled_tasks.router)
+
+    # Public read-only user leaderboard publish page (/api/public/leaderboard)
+    app.include_router(public_leaderboard.router)
+
+    # Public embed Q&A status polling (/api/public/embed/*)
+    app.include_router(public_embed.router)
+    app.include_router(public_config.router)
+    app.include_router(html_page_favorites.router)
+
+    # Product knowledge base API is mounted at /api/knowledge
+    app.include_router(knowledge.router)
+
+    # External knowledge ingest (publish md artifacts) at /api/knowledge-ingest
+    app.include_router(knowledge_ingest.router)
+
+    # System settings API is mounted at /api/system-settings
+    app.include_router(system_settings.router)
+
+    # Admin LLM metrics audit + export
+    app.include_router(llm_metrics.router)
+
+    # Admin window into other users' threads / messages / memory
+    app.include_router(admin_users.router)
+    app.include_router(admin_users.self_router)
+
+    app.include_router(admin_active_runs.router)
+
+    # Admin tool / skill call metrics audit + export
+    app.include_router(tool_metrics.router)
+
+    # Channels API is mounted at /api/channels
+    app.include_router(channels.router)
+
+    # Admin checkpoint migration API: SQLite legacy checkpoints -> Postgres.
+    app.include_router(checkpoint_migration.router)
+
+    # Assistants compatibility API (LangGraph Platform stub)
+    app.include_router(assistants_compat.router)
+
+    # Auth API is mounted at /api/v1/auth
+    app.include_router(auth.router)
+
+    # Feedback API is mounted at /api/threads/{thread_id}/runs/{run_id}/feedback
+    app.include_router(feedback.router)
+
+    # Notifications API is mounted at /api/notifications
+    app.include_router(notifications.router)
+
+    # Tags API is mounted at /api/tags
+    app.include_router(tags.router)
+
+    # 岗位 (position) management API is mounted at /api/positions
+    app.include_router(positions.router)
+
+    # Thread Runs API (LangGraph Platform-compatible runs lifecycle)
+    app.include_router(thread_runs.router)
+
+    app.include_router(thread_shares.router)
+    # Public, unauthenticated read-only share snapshots (/api/public/shares/*).
+    app.include_router(thread_shares.public_router)
+
+    # ?????????import ???? / view ????????
+    app.include_router(roundtable_draft_shares.router)
+    # Public, unauthenticated read-only roundtable result (/api/public/roundtable-shares/*).
+    app.include_router(roundtable_draft_shares.public_router)
+
+    # Stateless Runs API (stream/wait without a pre-existing thread)
+    app.include_router(runs.router)
+
+    # 开放问答接口 /api/open/chat —— 无需登录,指定模型 + 智能体做问答,可关闭思考模式
+    app.include_router(open_chat.router)
+
+    # 开放接口 /api/open/3qfx/ask —— 无需登录,按 taskId 直接发起 3qfx 第一次问答(后台运行,落成真实会话)
+    app.include_router(open_3qfx.router)
+
+    # --- ?????????? /page/workspace/roundtable/planning?---
+    # ??? app/gateway/routers/{multi_agent,intent,recommend}.py?
+    # ??? frontend-web/docs/multi-agent-backend-dev.md
+    app.include_router(multi_agent.router)   # Step 2?init(SSE) + run/stream(leader|special)
+    app.include_router(intent.router)        # Step 1?/api/intent/init + /stream
+    app.include_router(recommend.router)     # Step 1?2?/api/recommend/stream?? init?
+    # Step 3 ????/?????? artifacts.router?/api/threads/{id}/artifacts?
+
+    # Health probes are always unauthenticated (see
+    # auth_middleware.PUBLIC_HEALTH_PATHS). Support the bare Gateway path,
+    # API base URL, and LangGraph proxy base URL for deployment monitors.
+    @app.api_route("/health", methods=["GET", "HEAD"], tags=["health"])
+    @app.api_route("/api/health", methods=["GET", "HEAD"], tags=["health"])
+    @app.api_route("/api/langgraph/health", methods=["GET", "HEAD"], tags=["health"])
+    async def health_check() -> dict:
+        """Health check endpoint.
+
+        Returns:
+            Service health status information.
+        """
+        return {"status": "healthy", "service": "deer-flow-gateway"}
+
+    # WeKnora's embedded frontend uses root-relative routes such as /platform,
+    # /assets, /locales and /api/v1/*. Register that proxy last so it stays a
+    # true fallback and cannot intercept DeerFlow APIs.
+    # Register the public-prefix routes before the unprefixed static fallback.
+    # Otherwise proxy_router's /{proxied_path:path} route consumes every
+    # /deerflow/* request before the prefixed platform/assets routes can match.
+    app.include_router(llmwiki.proxy_router, prefix="/deerflow")
+    app.include_router(llmwiki.proxy_router)
+
+    return app
+
+
+# Create app instance for uvicorn
+app = create_app()
diff --git a/offline-backend-20260512/backend/app/gateway/deps.py b/offline-backend-20260512/backend/app/gateway/deps.py
index b094f05e..5a10142f 100644
--- a/offline-backend-20260512/backend/app/gateway/deps.py
+++ b/offline-backend-20260512/backend/app/gateway/deps.py
@@ -1,286 +1,286 @@
-"""Centralized accessors for singleton objects stored on ``app.state``.
-
-**Getters** (used by routers): raise 503 when a required dependency is
-missing, except ``get_store`` which returns ``None``.
-
-Initialization is handled directly in ``app.py`` via :class:`AsyncExitStack`.
-"""
-
-from __future__ import annotations
-
-import asyncio
-import logging
-from collections.abc import AsyncGenerator, Callable
-from contextlib import AsyncExitStack, asynccontextmanager
-from typing import TYPE_CHECKING, TypeVar, cast
-
-from fastapi import FastAPI, HTTPException, Request
-from langgraph.types import Checkpointer
-
-from deerflow.config.app_config import AppConfig
-from deerflow.persistence.feedback import FeedbackRepository
-from deerflow.runtime import RunContext, RunManager, StreamBridge
-from deerflow.runtime.events.store.base import RunEventStore
-from deerflow.runtime.runs.store.base import RunStore
-
-logger = logging.getLogger(__name__)
-
-if TYPE_CHECKING:
-    from app.gateway.auth.local_provider import LocalAuthProvider
-    from app.gateway.auth.repositories.sqlite import SQLiteUserRepository
-    from deerflow.persistence.agents import AgentStore
-    from deerflow.persistence.notifications import NotificationStore
-    from deerflow.persistence.scheduled_tasks import ScheduledTaskStore
-    from deerflow.persistence.skills import SkillStore
-    from deerflow.persistence.tags import TagStore
-    from deerflow.persistence.thread_meta.base import ThreadMetaStore
-
-
-T = TypeVar("T")
-
-
-def resolve_roundtable_gateway_mode(raw: str | None) -> str:
-    """解析 ``ROUNDTABLE_JOB_GATEWAY``：**默认 ``inprocess``（真模型后台挂起）**。
-
-    只有显式写 ``simulated`` 才用模拟网关（前端联调用：真实进度、不调真模型）；
-    缺省 / 空串 / 其它值一律落到 ``inprocess`` —— 部署时无需在 .env 里配置任何值，
-    后台挂起开箱即用（Phase 3b 已在 Windows 本地与内网 Docker 验证）。
-    """
-    mode = (raw or "").strip().lower()
-    return "simulated" if mode == "simulated" else "inprocess"
-
-
-def get_config(request: Request) -> AppConfig:
-    """Return the app-scoped ``AppConfig`` stored on ``app.state``."""
-    config = getattr(request.app.state, "config", None)
-    if config is None:
-        raise HTTPException(status_code=503, detail="Configuration not available")
-    return config
-
-
-@asynccontextmanager
-async def langgraph_runtime(app: FastAPI) -> AsyncGenerator[None, None]:
-    """Bootstrap and tear down all LangGraph runtime singletons.
-
-    Usage in ``app.py``::
-
-        async with langgraph_runtime(app):
-            yield
-    """
-    from deerflow.persistence.engine import close_engine, get_session_factory, init_engine_from_config
-    from deerflow.runtime import make_store, make_stream_bridge
-    from deerflow.runtime.checkpointer.async_provider import make_checkpointer
-    from deerflow.runtime.events.store import make_run_event_store
-
-    async with AsyncExitStack() as stack:
-        config = getattr(app.state, "config", None)
-        if config is None:
-            raise RuntimeError("langgraph_runtime() requires app.state.config to be initialized")
-
-        app.state.stream_bridge = await stack.enter_async_context(make_stream_bridge(config))
-
-        # Initialize persistence engine BEFORE checkpointer so that
-        # auto-create-database logic runs first (postgres backend).
-        await init_engine_from_config(config.database)
-
-        app.state.checkpointer = await stack.enter_async_context(make_checkpointer(config))
-        app.state.store = await stack.enter_async_context(make_store(config))
-
-        # Initialize repositories — one get_session_factory() call for all.
-        sf = get_session_factory()
-        if sf is not None:
-            from deerflow.persistence.feedback import FeedbackRepository
-            from deerflow.persistence.run import RunRepository
-
-            app.state.run_store = RunRepository(sf)
-            app.state.feedback_repo = FeedbackRepository(sf)
-        else:
-            from deerflow.runtime.runs.store.memory import MemoryRunStore
-
-            app.state.run_store = MemoryRunStore()
-            app.state.feedback_repo = None
-
-        from deerflow.persistence.thread_meta import make_thread_store
-
-        app.state.thread_store = make_thread_store(sf, app.state.store)
-
-        from deerflow.persistence.scheduled_tasks import make_scheduled_task_store
-
-        app.state.scheduled_task_store = make_scheduled_task_store(sf)
-
-        from deerflow.persistence.html_pages import make_html_page_store
-
-        app.state.html_page_store = make_html_page_store(sf)
-
-        from deerflow.persistence.llm_metrics import make_llm_metrics_store
-
-        app.state.llm_metrics_store = make_llm_metrics_store(sf)
-
-        from deerflow.persistence.concurrency import make_concurrency_sample_store
-
-        app.state.concurrency_sample_store = make_concurrency_sample_store(sf)
-
-        from deerflow.persistence.tool_metrics import make_tool_metrics_store
-
-        app.state.tool_metrics_store = make_tool_metrics_store(sf)
-
-        from deerflow.persistence.agents import make_agent_store
-
-        app.state.agent_store = make_agent_store(sf)
-
-        from deerflow.persistence.skills import make_skill_store
-
-        app.state.skill_store = make_skill_store(sf)
-
-        from deerflow.persistence.notifications import (
-            make_notification_store,
-            set_notification_store,
-        )
-
-        app.state.notification_store = make_notification_store(sf)
-        # Make the sender helper (harness layer) able to write without going
-        # through FastAPI request context — used by cascade.py.
-        set_notification_store(app.state.notification_store)
-        from deerflow.persistence.recommended_questions import make_recommended_question_store
-
-        app.state.recommended_question_store = make_recommended_question_store(sf)
-
-        from deerflow.persistence.tags import make_tag_store
-
-        app.state.tag_store = make_tag_store(sf)
-
-        from deerflow.persistence.positions import make_position_store
-
-        app.state.position_store = make_position_store(sf)
-
-        from deerflow.services.position_sync import PositionSyncService
-
-        app.state.position_sync = PositionSyncService(sf, app.state.position_store, app.state.tag_store)
-
-        from deerflow.persistence.thread_shares import make_thread_share_store
-
-        app.state.thread_share_store = make_thread_share_store(sf)
-
-        from deerflow.persistence.embed_sessions import make_embed_session_store
-
-        app.state.embed_session_store = make_embed_session_store(sf)
-
-        from deerflow.persistence.ai_writing_sessions import (
-            make_ai_writing_session_store,
-        )
-        from deerflow.persistence.ai_writing_sessions import (
-            register_default_store as _register_ai_writing_default_store,
-        )
-
-        app.state.ai_writing_session_store = make_ai_writing_session_store(sf)
-        # 让 LangGraph Server 路径下的 AI 写作节点能跨 app/harness 边界拿到 repo。
-        # 老 /start /stream /resume 路由仍走 request.app.state，互不影响。
-        _register_ai_writing_default_store(app.state.ai_writing_session_store)
-
-        # AI 写作「后台挂起」执行器：把写作图作为后台任务驱动到定稿（离页/刷新/切号不打断）。
-        # 用 app.state.checkpointer（与运行时 worker 同一份 saver）续上线程 checkpoint。
-        from app.gateway.ai_writing_job_executor import AIWritingJobExecutor
-
-        app.state.ai_writing_job_executor = AIWritingJobExecutor(app)
-        # 进程重启续跑：把仍标记为 background 的会话重新拉起驱动（best-effort，不阻塞启动）。
-        if app.state.ai_writing_session_store is not None:
-            from app.gateway.ai_writing_job_executor import STATUS_BACKGROUND
-
-            async def _reconcile_ai_writing_background() -> None:
-                try:
-                    sessions = await app.state.ai_writing_session_store.list_by_status(STATUS_BACKGROUND)
-                    app.state.ai_writing_job_executor.reconcile(sessions=sessions)
-                except Exception:  # noqa: BLE001
-                    logger.warning("[ai-writing-bg] startup reconcile failed", exc_info=True)
-
-            asyncio.create_task(_reconcile_ai_writing_background())
-
-        from deerflow.persistence.article_types import make_article_type_store
-
-        app.state.article_type_store = make_article_type_store(sf)
-
-        from deerflow.persistence.research_templates import make_research_template_store
-
-        app.state.research_template_store = make_research_template_store(sf)
-
-        from deerflow.persistence.memory_config import make_user_memory_config_store
-
-        app.state.memory_config_store = make_user_memory_config_store(sf)
-
-        from deerflow.knowledge import make_knowledge_service
-
-        app.state.knowledge_service = make_knowledge_service(sf, config.knowledge)
-
-        # Seed the built-in extraction prompts as editable templates so operators
-        # can immediately view/edit them (idempotent; never overwrites edits).
-        if app.state.knowledge_service is not None:
-            try:
-                await app.state.knowledge_service.seed_default_templates()
-            except Exception:
-                logger.exception("Failed to seed knowledge extraction templates")
-
-        # Background auto-sedimentation queue (phase 3) — drains finished
-        # conversations off the chat path. Worker only does work when
-        # ``knowledge.auto_ingest_enabled`` is set; the queue is always wired so
-        # the middleware can enqueue without a hard dependency.
-        from deerflow.knowledge.auto_ingest import AutoIngestQueue, set_auto_ingest_queue
-
-        if app.state.knowledge_service is not None:
-            _auto_ingest_queue = AutoIngestQueue(
-                lambda: getattr(app.state, "knowledge_service", None),
-                min_chars=config.knowledge.auto_ingest_min_chars,
-            )
-            _auto_ingest_queue.start()
-            set_auto_ingest_queue(_auto_ingest_queue)
-            app.state.knowledge_auto_ingest_queue = _auto_ingest_queue
-
-        from deerflow.persistence.user_prompts import make_user_prompt_store
-
-        app.state.user_prompt_store = make_user_prompt_store(sf)
-
-        from deerflow.persistence.user_preferences import make_user_preferences_store
-
-        app.state.user_preferences_store = make_user_preferences_store(sf)
-
-        from deerflow.persistence.roundtable_drafts import make_roundtable_draft_store
-
-        app.state.roundtable_draft_store = make_roundtable_draft_store(sf)
-
-        # Task-scoped roundtable drafts (taskId 深链聊天记录，独立存、不按 user 分权)。
-        from deerflow.persistence.roundtable_task_drafts import make_roundtable_task_draft_store
-
-        app.state.roundtable_task_draft_store = make_roundtable_task_draft_store(sf)
-
-        from deerflow.persistence.roundtable_draft_shares import make_roundtable_draft_share_store
-
-        app.state.roundtable_draft_share_store = make_roundtable_draft_share_store(sf)
-
-        from deerflow.persistence.roundtable_chains import make_roundtable_chain_store
-
-        app.state.roundtable_chain_store = make_roundtable_chain_store(sf)
-
-        # 独立的岗位协同 Session / Node 存储。它只索引业务流程状态，真实消息和
-        # 文件仍复用 LangGraph thread/checkpoint 与现有 artifact API。
-        from deerflow.persistence.position_roundtable import make_position_roundtable_store
-
-        app.state.position_roundtable_store = make_position_roundtable_store(sf)
-
-        # 岗位会商工作视角（名称 / 顺序 / 特殊意图岗）是全局配置，和组织/RBAC
-        # 岗位分开保存。启动只补齐缺失内置项，不覆盖管理员的既有修改。
-        from app.gateway.position_role_defaults import POSITION_ROLE_SEED
-        from deerflow.persistence.position_roles import make_position_role_store
-
-        app.state.position_role_store = make_position_role_store(sf)
-        await app.state.position_role_store.ensure_seed(POSITION_ROLE_SEED)
-
-        from deerflow.persistence.business_mapping import make_business_mapping_store
-
-        app.state.business_mapping_store = make_business_mapping_store(sf)
-
-        from deerflow.persistence.light_apps import make_light_app_store
-
-        app.state.light_app_store = make_light_app_store(sf)
-
+"""Centralized accessors for singleton objects stored on ``app.state``.
+
+**Getters** (used by routers): raise 503 when a required dependency is
+missing, except ``get_store`` which returns ``None``.
+
+Initialization is handled directly in ``app.py`` via :class:`AsyncExitStack`.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+from collections.abc import AsyncGenerator, Callable
+from contextlib import AsyncExitStack, asynccontextmanager
+from typing import TYPE_CHECKING, TypeVar, cast
+
+from fastapi import FastAPI, HTTPException, Request
+from langgraph.types import Checkpointer
+
+from deerflow.config.app_config import AppConfig
+from deerflow.persistence.feedback import FeedbackRepository
+from deerflow.runtime import RunContext, RunManager, StreamBridge
+from deerflow.runtime.events.store.base import RunEventStore
+from deerflow.runtime.runs.store.base import RunStore
+
+logger = logging.getLogger(__name__)
+
+if TYPE_CHECKING:
+    from app.gateway.auth.local_provider import LocalAuthProvider
+    from app.gateway.auth.repositories.sqlite import SQLiteUserRepository
+    from deerflow.persistence.agents import AgentStore
+    from deerflow.persistence.notifications import NotificationStore
+    from deerflow.persistence.scheduled_tasks import ScheduledTaskStore
+    from deerflow.persistence.skills import SkillStore
+    from deerflow.persistence.tags import TagStore
+    from deerflow.persistence.thread_meta.base import ThreadMetaStore
+
+
+T = TypeVar("T")
+
+
+def resolve_roundtable_gateway_mode(raw: str | None) -> str:
+    """解析 ``ROUNDTABLE_JOB_GATEWAY``：**默认 ``inprocess``（真模型后台挂起）**。
+
+    只有显式写 ``simulated`` 才用模拟网关（前端联调用：真实进度、不调真模型）；
+    缺省 / 空串 / 其它值一律落到 ``inprocess`` —— 部署时无需在 .env 里配置任何值，
+    后台挂起开箱即用（Phase 3b 已在 Windows 本地与内网 Docker 验证）。
+    """
+    mode = (raw or "").strip().lower()
+    return "simulated" if mode == "simulated" else "inprocess"
+
+
+def get_config(request: Request) -> AppConfig:
+    """Return the app-scoped ``AppConfig`` stored on ``app.state``."""
+    config = getattr(request.app.state, "config", None)
+    if config is None:
+        raise HTTPException(status_code=503, detail="Configuration not available")
+    return config
+
+
+@asynccontextmanager
+async def langgraph_runtime(app: FastAPI) -> AsyncGenerator[None, None]:
+    """Bootstrap and tear down all LangGraph runtime singletons.
+
+    Usage in ``app.py``::
+
+        async with langgraph_runtime(app):
+            yield
+    """
+    from deerflow.persistence.engine import close_engine, get_session_factory, init_engine_from_config
+    from deerflow.runtime import make_store, make_stream_bridge
+    from deerflow.runtime.checkpointer.async_provider import make_checkpointer
+    from deerflow.runtime.events.store import make_run_event_store
+
+    async with AsyncExitStack() as stack:
+        config = getattr(app.state, "config", None)
+        if config is None:
+            raise RuntimeError("langgraph_runtime() requires app.state.config to be initialized")
+
+        app.state.stream_bridge = await stack.enter_async_context(make_stream_bridge(config))
+
+        # Initialize persistence engine BEFORE checkpointer so that
+        # auto-create-database logic runs first (postgres backend).
+        await init_engine_from_config(config.database)
+
+        app.state.checkpointer = await stack.enter_async_context(make_checkpointer(config))
+        app.state.store = await stack.enter_async_context(make_store(config))
+
+        # Initialize repositories — one get_session_factory() call for all.
+        sf = get_session_factory()
+        if sf is not None:
+            from deerflow.persistence.feedback import FeedbackRepository
+            from deerflow.persistence.run import RunRepository
+
+            app.state.run_store = RunRepository(sf)
+            app.state.feedback_repo = FeedbackRepository(sf)
+        else:
+            from deerflow.runtime.runs.store.memory import MemoryRunStore
+
+            app.state.run_store = MemoryRunStore()
+            app.state.feedback_repo = None
+
+        from deerflow.persistence.thread_meta import make_thread_store
+
+        app.state.thread_store = make_thread_store(sf, app.state.store)
+
+        from deerflow.persistence.scheduled_tasks import make_scheduled_task_store
+
+        app.state.scheduled_task_store = make_scheduled_task_store(sf)
+
+        from deerflow.persistence.html_pages import make_html_page_store
+
+        app.state.html_page_store = make_html_page_store(sf)
+
+        from deerflow.persistence.llm_metrics import make_llm_metrics_store
+
+        app.state.llm_metrics_store = make_llm_metrics_store(sf)
+
+        from deerflow.persistence.concurrency import make_concurrency_sample_store
+
+        app.state.concurrency_sample_store = make_concurrency_sample_store(sf)
+
+        from deerflow.persistence.tool_metrics import make_tool_metrics_store
+
+        app.state.tool_metrics_store = make_tool_metrics_store(sf)
+
+        from deerflow.persistence.agents import make_agent_store
+
+        app.state.agent_store = make_agent_store(sf)
+
+        from deerflow.persistence.skills import make_skill_store
+
+        app.state.skill_store = make_skill_store(sf)
+
+        from deerflow.persistence.notifications import (
+            make_notification_store,
+            set_notification_store,
+        )
+
+        app.state.notification_store = make_notification_store(sf)
+        # Make the sender helper (harness layer) able to write without going
+        # through FastAPI request context — used by cascade.py.
+        set_notification_store(app.state.notification_store)
+        from deerflow.persistence.recommended_questions import make_recommended_question_store
+
+        app.state.recommended_question_store = make_recommended_question_store(sf)
+
+        from deerflow.persistence.tags import make_tag_store
+
+        app.state.tag_store = make_tag_store(sf)
+
+        from deerflow.persistence.positions import make_position_store
+
+        app.state.position_store = make_position_store(sf)
+
+        from deerflow.services.position_sync import PositionSyncService
+
+        app.state.position_sync = PositionSyncService(sf, app.state.position_store, app.state.tag_store)
+
+        from deerflow.persistence.thread_shares import make_thread_share_store
+
+        app.state.thread_share_store = make_thread_share_store(sf)
+
+        from deerflow.persistence.embed_sessions import make_embed_session_store
+
+        app.state.embed_session_store = make_embed_session_store(sf)
+
+        from deerflow.persistence.ai_writing_sessions import (
+            make_ai_writing_session_store,
+        )
+        from deerflow.persistence.ai_writing_sessions import (
+            register_default_store as _register_ai_writing_default_store,
+        )
+
+        app.state.ai_writing_session_store = make_ai_writing_session_store(sf)
+        # 让 LangGraph Server 路径下的 AI 写作节点能跨 app/harness 边界拿到 repo。
+        # 老 /start /stream /resume 路由仍走 request.app.state，互不影响。
+        _register_ai_writing_default_store(app.state.ai_writing_session_store)
+
+        # AI 写作「后台挂起」执行器：把写作图作为后台任务驱动到定稿（离页/刷新/切号不打断）。
+        # 用 app.state.checkpointer（与运行时 worker 同一份 saver）续上线程 checkpoint。
+        from app.gateway.ai_writing_job_executor import AIWritingJobExecutor
+
+        app.state.ai_writing_job_executor = AIWritingJobExecutor(app)
+        # 进程重启续跑：把仍标记为 background 的会话重新拉起驱动（best-effort，不阻塞启动）。
+        if app.state.ai_writing_session_store is not None:
+            from app.gateway.ai_writing_job_executor import STATUS_BACKGROUND
+
+            async def _reconcile_ai_writing_background() -> None:
+                try:
+                    sessions = await app.state.ai_writing_session_store.list_by_status(STATUS_BACKGROUND)
+                    app.state.ai_writing_job_executor.reconcile(sessions=sessions)
+                except Exception:  # noqa: BLE001
+                    logger.warning("[ai-writing-bg] startup reconcile failed", exc_info=True)
+
+            asyncio.create_task(_reconcile_ai_writing_background())
+
+        from deerflow.persistence.article_types import make_article_type_store
+
+        app.state.article_type_store = make_article_type_store(sf)
+
+        from deerflow.persistence.research_templates import make_research_template_store
+
+        app.state.research_template_store = make_research_template_store(sf)
+
+        from deerflow.persistence.memory_config import make_user_memory_config_store
+
+        app.state.memory_config_store = make_user_memory_config_store(sf)
+
+        from deerflow.knowledge import make_knowledge_service
+
+        app.state.knowledge_service = make_knowledge_service(sf, config.knowledge)
+
+        # Seed the built-in extraction prompts as editable templates so operators
+        # can immediately view/edit them (idempotent; never overwrites edits).
+        if app.state.knowledge_service is not None:
+            try:
+                await app.state.knowledge_service.seed_default_templates()
+            except Exception:
+                logger.exception("Failed to seed knowledge extraction templates")
+
+        # Background auto-sedimentation queue (phase 3) — drains finished
+        # conversations off the chat path. Worker only does work when
+        # ``knowledge.auto_ingest_enabled`` is set; the queue is always wired so
+        # the middleware can enqueue without a hard dependency.
+        from deerflow.knowledge.auto_ingest import AutoIngestQueue, set_auto_ingest_queue
+
+        if app.state.knowledge_service is not None:
+            _auto_ingest_queue = AutoIngestQueue(
+                lambda: getattr(app.state, "knowledge_service", None),
+                min_chars=config.knowledge.auto_ingest_min_chars,
+            )
+            _auto_ingest_queue.start()
+            set_auto_ingest_queue(_auto_ingest_queue)
+            app.state.knowledge_auto_ingest_queue = _auto_ingest_queue
+
+        from deerflow.persistence.user_prompts import make_user_prompt_store
+
+        app.state.user_prompt_store = make_user_prompt_store(sf)
+
+        from deerflow.persistence.user_preferences import make_user_preferences_store
+
+        app.state.user_preferences_store = make_user_preferences_store(sf)
+
+        from deerflow.persistence.roundtable_drafts import make_roundtable_draft_store
+
+        app.state.roundtable_draft_store = make_roundtable_draft_store(sf)
+
+        # Task-scoped roundtable drafts (taskId 深链聊天记录，独立存、不按 user 分权)。
+        from deerflow.persistence.roundtable_task_drafts import make_roundtable_task_draft_store
+
+        app.state.roundtable_task_draft_store = make_roundtable_task_draft_store(sf)
+
+        from deerflow.persistence.roundtable_draft_shares import make_roundtable_draft_share_store
+
+        app.state.roundtable_draft_share_store = make_roundtable_draft_share_store(sf)
+
+        from deerflow.persistence.roundtable_chains import make_roundtable_chain_store
+
+        app.state.roundtable_chain_store = make_roundtable_chain_store(sf)
+
+        # 独立的岗位协同 Session / Node 存储。它只索引业务流程状态，真实消息和
+        # 文件仍复用 LangGraph thread/checkpoint 与现有 artifact API。
+        from deerflow.persistence.position_roundtable import make_position_roundtable_store
+
+        app.state.position_roundtable_store = make_position_roundtable_store(sf)
+
+        # 岗位会商工作视角（名称 / 顺序 / 特殊意图岗）是全局配置，和组织/RBAC
+        # 岗位分开保存。启动只补齐缺失内置项，不覆盖管理员的既有修改。
+        from app.gateway.position_role_defaults import POSITION_ROLE_SEED
+        from deerflow.persistence.position_roles import make_position_role_store
+
+        app.state.position_role_store = make_position_role_store(sf)
+        await app.state.position_role_store.ensure_seed(POSITION_ROLE_SEED)
+
+        from deerflow.persistence.business_mapping import make_business_mapping_store
+
+        app.state.business_mapping_store = make_business_mapping_store(sf)
+
+        from deerflow.persistence.light_apps import make_light_app_store
+
+        app.state.light_app_store = make_light_app_store(sf)
+
         from deerflow.persistence.menu_overrides import make_menu_override_store
 
         app.state.menu_override_store = make_menu_override_store(sf)
@@ -290,180 +290,184 @@ async def langgraph_runtime(app: FastAPI) -> AsyncGenerator[None, None]:
         app.state.compact_menu_override_store = make_compact_menu_override_store(sf)
 
         from deerflow.persistence.sensitive_words import make_sensitive_word_store
-
-        app.state.sensitive_word_store = make_sensitive_word_store(sf)
-
-        from deerflow.persistence.dashboard_sessions import make_dashboard_session_store
-
-        app.state.dashboard_session_store = make_dashboard_session_store(sf)
-
-        from deerflow.persistence.task_buttons import make_task_button_store
-
-        app.state.task_button_store = make_task_button_store(sf)
-
-        from deerflow.integrations.weknora.runtime import register_llmwiki_store
-        from deerflow.persistence.llmwiki import make_llmwiki_store
-
-        app.state.llmwiki_store = make_llmwiki_store(sf)
-        register_llmwiki_store(app.state.llmwiki_store)
-
-        from deerflow.persistence.taskcop_tasks import make_taskcop_task_store
-
-        app.state.taskcop_task_store = make_taskcop_task_store(sf)
-
-        from deerflow.persistence.task_reports import make_task_report_store
-
-        app.state.task_report_store = make_task_report_store(sf)
-
-        from deerflow.persistence.roundtable_artifact_submissions import make_roundtable_artifact_submission_store
-
-        # Task-scoped artifact submission status: shared by task; actor is audit-only.
-        app.state.roundtable_artifact_submission_store = make_roundtable_artifact_submission_store(sf)
-
-        from deerflow.persistence.parallel_agent_tasks import make_parallel_agent_task_store
-
-        app.state.parallel_agent_task_store = make_parallel_agent_task_store(sf)
-
-        from deerflow.persistence.roundtable_jobs import (
-            make_roundtable_job_store,
-        )
-        from deerflow.persistence.roundtable_jobs import (
-            register_default_store as _register_roundtable_job_default_store,
-        )
-
-        app.state.roundtable_job_store = make_roundtable_job_store(sf)
-        # 让 LangGraph Server 路径下的圆桌编排图节点能跨 app/harness 边界拿到 repo
-        # （Phase 2 后台编排图写进度用）。
-        _register_roundtable_job_default_store(app.state.roundtable_job_store)
-
-        # 意图回合单飞 store（Phase 4）：``/api/intent/stream`` 据此对 (thread_id,
-        # client_turn_id) 跨 worker 单飞 + 幂等回放，杜绝并发双发意图。sf 为 None（无 DB）
-        # 时为 None，路由侧判空跳过单飞、走原流程。
-        from deerflow.persistence.intent_turns import make_intent_turn_store
-
-        app.state.intent_turn_store = make_intent_turn_store(sf)
-
-        # 圆桌会商「诊断日志」store：会商全流程各处报错 / 关键事件落库，供管理员诊断页筛选。
-        # 同样登记进程级默认仓储——harness 侧编排引擎 / 进程内网关跨边界写诊断用。
-        from deerflow.persistence.roundtable_diagnostics import (
-            make_roundtable_diagnostic_store,
-        )
-        from deerflow.persistence.roundtable_diagnostics import (
-            register_default_store as _register_roundtable_diagnostic_default_store,
-        )
-
-        app.state.roundtable_diagnostic_store = make_roundtable_diagnostic_store(sf)
-        _register_roundtable_diagnostic_default_store(app.state.roundtable_diagnostic_store)
-
-        # 圆桌后台作业执行器：把编排引擎作为后台任务跑。
-        # 网关由环境变量 ``ROUNDTABLE_JOB_GATEWAY`` 选择（见 resolve_roundtable_gateway_mode）：
-        #   - 缺省 / "inprocess"（Phase 3b）：进程内驱动真模型（leader/seat/report）——**默认**，
-        #     部署时无需任何配置即可用后台挂起；
-        #   - 显式 "simulated"（Phase 3a）：模拟网关，产生真实进度但不调真模型，供前端联调。
-        if app.state.roundtable_job_store is not None:
-            import os
-
-            from app.gateway.roundtable_job_executor import RoundtableJobExecutor
-
-            gateway_mode = resolve_roundtable_gateway_mode(os.getenv("ROUNDTABLE_JOB_GATEWAY"))
-            if gateway_mode == "simulated":
-                from deerflow.agents.roundtable_orchestrator import SimulatedRoundtableGateway
-
-                # delay=3s/步：让模拟后台任务有可观察的进行中时长（切页/重连仍在跑）。
-                gateway_factory = lambda p: SimulatedRoundtableGateway(  # noqa: E731
-                    p.agents, mode=p.mode, delay=3.0
-                )
-                logger.info("[roundtable-jobs] gateway = simulated")
-            else:
-                from app.gateway.roundtable_inprocess_gateway import InProcessRoundtableGateway
-
-                _app = app  # 闭包捕获 app，供网关访问 app.state（checkpointer/thread_store）
-                gateway_factory = lambda p: InProcessRoundtableGateway(  # noqa: E731
-                    _app,
-                    user_id=p.user_id,
-                    agents=p.agents,
-                    model=p.model,
-                    # 席位执行模式沿用 Step2 选择，后台席位 run 据此覆盖 policy 的 thinking/reasoning + 开子代理。
-                    seat_thinking_enabled=p.seat_thinking_enabled,
-                    seat_reasoning_effort=p.seat_reasoning_effort,
-                    seat_subagent_enabled=p.seat_subagent_enabled,
-                    seat_skill_directives=p.seat_skill_directives,
-                    # 每席位推理深度覆盖（业务链条里各席位单独配）：run_seat 命中即覆盖作业级 seatMode。
-                    seat_modes=p.seat_modes,
-                    # 诊断日志上下文：网关每轮 leader/seat/report/summary/dashboard 报错均带这些归属。
-                    job_id=p.job_id,
-                    draft_id=p.draft_id,
-                    task_id=p.task_id,
-                )
-                logger.info("[roundtable-jobs] gateway = inprocess (real models, default)")
-
-            app.state.roundtable_job_executor = RoundtableJobExecutor(
-                app.state.roundtable_job_store,
-                gateway_factory=gateway_factory,
-                draft_store=getattr(app.state, "roundtable_draft_store", None),
-                # task 作业把最终报告写回**独立的** task-draft 存储（不分权）。
-                task_draft_store=getattr(app.state, "roundtable_task_draft_store", None),
-            )
-
-            # Phase 3 dispatcher：跨 worker 领取 queued/租约过期作业 + outbox 补偿器。
-            # 每个 worker 进程都跑一个；谁先 claim_job 成功谁执行（DB 条件 UPDATE 互斥），
-            # 崩溃的 worker 租约过期后由其它 worker 接管。路由 start/resume 后 nudge 立即唤醒。
-            from app.gateway.roundtable_job_dispatcher import RoundtableJobDispatcher
-
-            app.state.roundtable_job_dispatcher = RoundtableJobDispatcher(
-                app.state.roundtable_job_store,
-                app.state.roundtable_job_executor,
-                draft_store=getattr(app.state, "roundtable_draft_store", None),
-                task_draft_store=getattr(app.state, "roundtable_task_draft_store", None),
-            )
-            app.state.roundtable_job_dispatcher.start()
-        else:
-            app.state.roundtable_job_executor = None
-            app.state.roundtable_job_dispatcher = None
-
-        # ── Deep Research (深度研究) stores + executor + dispatcher ──────────
-        # Five tables: sessions / jobs / events / sources / messages.
-        # When DB is available, the executor + dispatcher run background research
-        # jobs (lease-based claim, crash-recoverable) mirroring the roundtable pattern.
-        from deerflow.persistence.deep_research_events import (
-            make_deep_research_event_store,
-        )
-        from deerflow.persistence.deep_research_jobs import make_deep_research_job_store
-        from deerflow.persistence.deep_research_messages import (
-            make_deep_research_message_store,
-        )
-        from deerflow.persistence.deep_research_sessions import (
-            make_deep_research_session_store,
-        )
+
+        app.state.sensitive_word_store = make_sensitive_word_store(sf)
+
+        from deerflow.persistence.dashboard_sessions import make_dashboard_session_store
+
+        app.state.dashboard_session_store = make_dashboard_session_store(sf)
+
+        from deerflow.persistence.task_buttons import make_task_button_store
+
+        app.state.task_button_store = make_task_button_store(sf)
+
+        from deerflow.persistence.report_structures import make_report_structure_store
+
+        app.state.report_structure_store = make_report_structure_store(sf)
+
+        from deerflow.integrations.weknora.runtime import register_llmwiki_store
+        from deerflow.persistence.llmwiki import make_llmwiki_store
+
+        app.state.llmwiki_store = make_llmwiki_store(sf)
+        register_llmwiki_store(app.state.llmwiki_store)
+
+        from deerflow.persistence.taskcop_tasks import make_taskcop_task_store
+
+        app.state.taskcop_task_store = make_taskcop_task_store(sf)
+
+        from deerflow.persistence.task_reports import make_task_report_store
+
+        app.state.task_report_store = make_task_report_store(sf)
+
+        from deerflow.persistence.roundtable_artifact_submissions import make_roundtable_artifact_submission_store
+
+        # Task-scoped artifact submission status: shared by task; actor is audit-only.
+        app.state.roundtable_artifact_submission_store = make_roundtable_artifact_submission_store(sf)
+
+        from deerflow.persistence.parallel_agent_tasks import make_parallel_agent_task_store
+
+        app.state.parallel_agent_task_store = make_parallel_agent_task_store(sf)
+
+        from deerflow.persistence.roundtable_jobs import (
+            make_roundtable_job_store,
+        )
+        from deerflow.persistence.roundtable_jobs import (
+            register_default_store as _register_roundtable_job_default_store,
+        )
+
+        app.state.roundtable_job_store = make_roundtable_job_store(sf)
+        # 让 LangGraph Server 路径下的圆桌编排图节点能跨 app/harness 边界拿到 repo
+        # （Phase 2 后台编排图写进度用）。
+        _register_roundtable_job_default_store(app.state.roundtable_job_store)
+
+        # 意图回合单飞 store（Phase 4）：``/api/intent/stream`` 据此对 (thread_id,
+        # client_turn_id) 跨 worker 单飞 + 幂等回放，杜绝并发双发意图。sf 为 None（无 DB）
+        # 时为 None，路由侧判空跳过单飞、走原流程。
+        from deerflow.persistence.intent_turns import make_intent_turn_store
+
+        app.state.intent_turn_store = make_intent_turn_store(sf)
+
+        # 圆桌会商「诊断日志」store：会商全流程各处报错 / 关键事件落库，供管理员诊断页筛选。
+        # 同样登记进程级默认仓储——harness 侧编排引擎 / 进程内网关跨边界写诊断用。
+        from deerflow.persistence.roundtable_diagnostics import (
+            make_roundtable_diagnostic_store,
+        )
+        from deerflow.persistence.roundtable_diagnostics import (
+            register_default_store as _register_roundtable_diagnostic_default_store,
+        )
+
+        app.state.roundtable_diagnostic_store = make_roundtable_diagnostic_store(sf)
+        _register_roundtable_diagnostic_default_store(app.state.roundtable_diagnostic_store)
+
+        # 圆桌后台作业执行器：把编排引擎作为后台任务跑。
+        # 网关由环境变量 ``ROUNDTABLE_JOB_GATEWAY`` 选择（见 resolve_roundtable_gateway_mode）：
+        #   - 缺省 / "inprocess"（Phase 3b）：进程内驱动真模型（leader/seat/report）——**默认**，
+        #     部署时无需任何配置即可用后台挂起；
+        #   - 显式 "simulated"（Phase 3a）：模拟网关，产生真实进度但不调真模型，供前端联调。
+        if app.state.roundtable_job_store is not None:
+            import os
+
+            from app.gateway.roundtable_job_executor import RoundtableJobExecutor
+
+            gateway_mode = resolve_roundtable_gateway_mode(os.getenv("ROUNDTABLE_JOB_GATEWAY"))
+            if gateway_mode == "simulated":
+                from deerflow.agents.roundtable_orchestrator import SimulatedRoundtableGateway
+
+                # delay=3s/步：让模拟后台任务有可观察的进行中时长（切页/重连仍在跑）。
+                gateway_factory = lambda p: SimulatedRoundtableGateway(  # noqa: E731
+                    p.agents, mode=p.mode, delay=3.0
+                )
+                logger.info("[roundtable-jobs] gateway = simulated")
+            else:
+                from app.gateway.roundtable_inprocess_gateway import InProcessRoundtableGateway
+
+                _app = app  # 闭包捕获 app，供网关访问 app.state（checkpointer/thread_store）
+                gateway_factory = lambda p: InProcessRoundtableGateway(  # noqa: E731
+                    _app,
+                    user_id=p.user_id,
+                    agents=p.agents,
+                    model=p.model,
+                    # 席位执行模式沿用 Step2 选择，后台席位 run 据此覆盖 policy 的 thinking/reasoning + 开子代理。
+                    seat_thinking_enabled=p.seat_thinking_enabled,
+                    seat_reasoning_effort=p.seat_reasoning_effort,
+                    seat_subagent_enabled=p.seat_subagent_enabled,
+                    seat_skill_directives=p.seat_skill_directives,
+                    # 每席位推理深度覆盖（业务链条里各席位单独配）：run_seat 命中即覆盖作业级 seatMode。
+                    seat_modes=p.seat_modes,
+                    # 诊断日志上下文：网关每轮 leader/seat/report/summary/dashboard 报错均带这些归属。
+                    job_id=p.job_id,
+                    draft_id=p.draft_id,
+                    task_id=p.task_id,
+                )
+                logger.info("[roundtable-jobs] gateway = inprocess (real models, default)")
+
+            app.state.roundtable_job_executor = RoundtableJobExecutor(
+                app.state.roundtable_job_store,
+                gateway_factory=gateway_factory,
+                draft_store=getattr(app.state, "roundtable_draft_store", None),
+                # task 作业把最终报告写回**独立的** task-draft 存储（不分权）。
+                task_draft_store=getattr(app.state, "roundtable_task_draft_store", None),
+            )
+
+            # Phase 3 dispatcher：跨 worker 领取 queued/租约过期作业 + outbox 补偿器。
+            # 每个 worker 进程都跑一个；谁先 claim_job 成功谁执行（DB 条件 UPDATE 互斥），
+            # 崩溃的 worker 租约过期后由其它 worker 接管。路由 start/resume 后 nudge 立即唤醒。
+            from app.gateway.roundtable_job_dispatcher import RoundtableJobDispatcher
+
+            app.state.roundtable_job_dispatcher = RoundtableJobDispatcher(
+                app.state.roundtable_job_store,
+                app.state.roundtable_job_executor,
+                draft_store=getattr(app.state, "roundtable_draft_store", None),
+                task_draft_store=getattr(app.state, "roundtable_task_draft_store", None),
+            )
+            app.state.roundtable_job_dispatcher.start()
+        else:
+            app.state.roundtable_job_executor = None
+            app.state.roundtable_job_dispatcher = None
+
+        # ── Deep Research (深度研究) stores + executor + dispatcher ──────────
+        # Five tables: sessions / jobs / events / sources / messages.
+        # When DB is available, the executor + dispatcher run background research
+        # jobs (lease-based claim, crash-recoverable) mirroring the roundtable pattern.
+        from deerflow.persistence.deep_research_events import (
+            make_deep_research_event_store,
+        )
+        from deerflow.persistence.deep_research_jobs import make_deep_research_job_store
+        from deerflow.persistence.deep_research_messages import (
+            make_deep_research_message_store,
+        )
+        from deerflow.persistence.deep_research_sessions import (
+            make_deep_research_session_store,
+        )
         from deerflow.persistence.deep_research_sources import (
             make_deep_research_source_store,
         )
         from deerflow.persistence.document_rewrite_versions import (
             make_document_rewrite_version_store,
         )
-
-        app.state.deep_research_session_store = make_deep_research_session_store(sf)
-        app.state.deep_research_job_store = make_deep_research_job_store(sf)
-        app.state.deep_research_event_store = make_deep_research_event_store(sf)
-        app.state.deep_research_source_store = make_deep_research_source_store(sf)
+
+        app.state.deep_research_session_store = make_deep_research_session_store(sf)
+        app.state.deep_research_job_store = make_deep_research_job_store(sf)
+        app.state.deep_research_event_store = make_deep_research_event_store(sf)
+        app.state.deep_research_source_store = make_deep_research_source_store(sf)
         app.state.deep_research_message_store = make_deep_research_message_store(sf)
         # Whole-document AI rewrites retain a durable original snapshot so a
         # completed replacement can be safely undone after a refresh.
         app.state.document_rewrite_version_store = make_document_rewrite_version_store(sf)
-        # Low-latency in-process SSE fan-out. The database event log remains
-        # authoritative for reconnect replay and cross-worker recovery.
-        from app.gateway.deep_research_live_hub import DeepResearchLiveHub
-
-        app.state.deep_research_live_hub = DeepResearchLiveHub()
-
-        if app.state.deep_research_job_store is not None:
-            from app.gateway.deep_research_job_dispatcher import (
-                DeepResearchJobDispatcher,
-            )
-            from app.gateway.deep_research_job_executor import (
-                DeepResearchJobExecutor,
-            )
-
+        # Low-latency in-process SSE fan-out. The database event log remains
+        # authoritative for reconnect replay and cross-worker recovery.
+        from app.gateway.deep_research_live_hub import DeepResearchLiveHub
+
+        app.state.deep_research_live_hub = DeepResearchLiveHub()
+
+        if app.state.deep_research_job_store is not None:
+            from app.gateway.deep_research_job_dispatcher import (
+                DeepResearchJobDispatcher,
+            )
+            from app.gateway.deep_research_job_executor import (
+                DeepResearchJobExecutor,
+            )
+
             app.state.deep_research_executor = DeepResearchJobExecutor(
                 app.state.deep_research_job_store,
                 app.state.deep_research_session_store,
@@ -474,358 +478,358 @@ async def langgraph_runtime(app: FastAPI) -> AsyncGenerator[None, None]:
                 document_rewrite_version_store=app.state.document_rewrite_version_store,
                 app_config=config,
             )
-            app.state.deep_research_dispatcher = DeepResearchJobDispatcher(
-                app.state.deep_research_job_store,
-                app.state.deep_research_executor,
-            )
-            app.state.deep_research_dispatcher.start()
-        else:
-            app.state.deep_research_executor = None
-            app.state.deep_research_dispatcher = None
-
-        # Run event store (has its own factory with config-driven backend selection)
-        run_events_config = getattr(config, "run_events", None)
-        app.state.run_event_store = make_run_event_store(run_events_config)
-
-        # RunManager with store backing for persistence
-        app.state.run_manager = RunManager(store=app.state.run_store)
-        from deerflow.runtime import make_redis_concurrency_gate
-
-        app.state.concurrency_gate = make_redis_concurrency_gate(config)
-
-        try:
-            yield
-        finally:
-            from deerflow.persistence.notifications import set_notification_store
-
-            set_notification_store(None)
-            dispatcher = getattr(app.state, "roundtable_job_dispatcher", None)
-            if dispatcher is not None:
-                await dispatcher.stop()
-            dr_dispatcher = getattr(app.state, "deep_research_dispatcher", None)
-            if dr_dispatcher is not None:
-                await dr_dispatcher.stop()
-            queue = getattr(app.state, "knowledge_auto_ingest_queue", None)
-            if queue is not None:
-                from deerflow.knowledge.auto_ingest import set_auto_ingest_queue
-
-                await queue.stop()
-                set_auto_ingest_queue(None)
-            await close_engine()
-            gate = getattr(app.state, "concurrency_gate", None)
-            if gate is not None:
-                await gate.aclose()
-
-
-# ---------------------------------------------------------------------------
-# Getters – called by routers per-request
-# ---------------------------------------------------------------------------
-
-
-def _require(attr: str, label: str) -> Callable[[Request], T]:
-    """Create a FastAPI dependency that returns ``app.state.<attr>`` or 503."""
-
-    def dep(request: Request) -> T:
-        val = getattr(request.app.state, attr, None)
-        if val is None:
-            raise HTTPException(status_code=503, detail=f"{label} not available")
-        return cast(T, val)
-
-    dep.__name__ = dep.__qualname__ = f"get_{attr}"
-    return dep
-
-
-get_stream_bridge: Callable[[Request], StreamBridge] = _require("stream_bridge", "Stream bridge")
-get_run_manager: Callable[[Request], RunManager] = _require("run_manager", "Run manager")
-get_checkpointer: Callable[[Request], Checkpointer] = _require("checkpointer", "Checkpointer")
-get_run_event_store: Callable[[Request], RunEventStore] = _require("run_event_store", "Run event store")
-get_feedback_repo: Callable[[Request], FeedbackRepository] = _require("feedback_repo", "Feedback")
-get_run_store: Callable[[Request], RunStore] = _require("run_store", "Run store")
-
-
-def get_concurrency_gate(request: Request):
-    """Return the optional cross-worker concurrency gate."""
-    return getattr(request.app.state, "concurrency_gate", None)
-
-
-def get_store(request: Request):
-    """Return the global store (may be ``None`` if not configured)."""
-    return getattr(request.app.state, "store", None)
-
-
-def get_thread_store(request: Request) -> ThreadMetaStore:
-    """Return the thread metadata store (SQL or memory-backed)."""
-    val = getattr(request.app.state, "thread_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Thread metadata store not available")
-    return val
-
-
-def get_scheduled_task_store(request: Request) -> ScheduledTaskStore:
-    """Return the scheduled-task store."""
-    val = getattr(request.app.state, "scheduled_task_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Scheduled task store not available")
-    return val
-
-
-def get_llm_metrics_store(request: Request):
-    """Return the LLM metrics store, or None when not configured."""
-    return getattr(request.app.state, "llm_metrics_store", None)
-
-
-def get_concurrency_sample_store(request: Request):
-    """Return the concurrency-sample store, or None on the memory backend."""
-    return getattr(request.app.state, "concurrency_sample_store", None)
-
-
-def get_tool_metrics_store(request: Request):
-    """Return the tool-call metrics store, or None when not configured."""
-    return getattr(request.app.state, "tool_metrics_store", None)
-
-
-def get_agent_store(request: Request) -> AgentStore:
-    """Return the custom-agent store."""
-    val = getattr(request.app.state, "agent_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Agent store not available")
-    return val
-
-
-def get_skill_store(request: Request) -> SkillStore:
-    """Return the custom-skill store."""
-    val = getattr(request.app.state, "skill_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Skill store not available")
-    return val
-
-
-def get_notification_store(request: Request) -> NotificationStore:
-    """Return the in-app notification store."""
-    val = getattr(request.app.state, "notification_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Notification store not available")
-    return val
-
-
-def get_tag_store(request: Request) -> TagStore:
-    """Return the tag store."""
-    val = getattr(request.app.state, "tag_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Tag store not available")
-    return val
-
-
-def get_position_store(request: Request):
-    """Return the 岗位 (position) store."""
-    val = getattr(request.app.state, "position_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Position store not available")
-    return val
-
-
-def get_position_sync(request: Request):
-    """Return the position → subscription sync service."""
-    val = getattr(request.app.state, "position_sync", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Position sync service not available")
-    return val
-
-
-def get_thread_share_store(request: Request):
-    """Return the conversation share-link store."""
-    val = getattr(request.app.state, "thread_share_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Thread share store not available")
-    return val
-
-
-def get_roundtable_draft_share_store(request: Request):
-    """Return the roundtable-draft share-link store."""
-    val = getattr(request.app.state, "roundtable_draft_share_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Roundtable draft share store not available")
-    return val
-
-
-def get_embed_session_store(request: Request):
-    """Return the iframe embed session → thread mapping store."""
-    val = getattr(request.app.state, "embed_session_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Embed session store not available")
-    return val
-
-
-def get_ai_writing_session_store(request: Request):
-    """Return the AI writing session store, or None when not configured."""
-    return getattr(request.app.state, "ai_writing_session_store", None)
-
-
-def get_ai_writing_job_executor(request: Request):
-    """Return the AI writing background-suspend executor, or None when unavailable."""
-    return getattr(request.app.state, "ai_writing_job_executor", None)
-
-
-def get_article_type_store(request: Request):
-    """Return the article type store, or None when not configured."""
-    return getattr(request.app.state, "article_type_store", None)
-
-
-def get_research_template_store(request: Request):
-    """Return the research template store, or None when not configured."""
-    return getattr(request.app.state, "research_template_store", None)
-
-
-def get_memory_config_store(request: Request):
-    """Return the per-user memory config store."""
-    val = getattr(request.app.state, "memory_config_store", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Memory config store not available")
-    return val
-
-
-def get_knowledge_service(request: Request):
-    """Return the knowledge base service (503 when disabled / memory backend)."""
-    val = getattr(request.app.state, "knowledge_service", None)
-    if val is None:
-        raise HTTPException(status_code=503, detail="Knowledge base not available")
-    return val
-
-
-def get_run_context(request: Request) -> RunContext:
-    """Build a :class:`RunContext` from ``app.state`` singletons.
-
-    Returns a *base* context with infrastructure dependencies.
-    """
-    config = get_config(request)
-    return RunContext(
-        checkpointer=get_checkpointer(request),
-        store=get_store(request),
-        event_store=get_run_event_store(request),
-        run_events_config=getattr(config, "run_events", None),
-        thread_store=get_thread_store(request),
-        app_config=config,
-    )
-
-
-# ---------------------------------------------------------------------------
-# Auth helpers (used by authz.py and auth middleware)
-# ---------------------------------------------------------------------------
-
-# Cached singletons to avoid repeated instantiation per request
-_cached_local_provider: LocalAuthProvider | None = None
-_cached_repo: SQLiteUserRepository | None = None
-
-
-def get_request_access_token(request: Request) -> str | None:
-    """Extract access token from cookie or Authorization header.
-
-    Priority:
-    1. ``access_token`` cookie (browser session mode)
-    2. ``Authorization: Bearer <token>`` header (API client mode)
-    3. ``access_token`` query parameter for the WeKnora iframe bootstrap only
-    """
-    cookie_token = request.cookies.get("access_token")
-    if cookie_token:
-        return cookie_token
-
-    authorization = request.headers.get("authorization", "").strip()
-    if authorization:
-        scheme, _, token = authorization.partition(" ")
-        if scheme.lower() == "bearer":
-            token = token.strip()
-            if token:
-                return token
-
-    from app.gateway.proxy_path import app_relative_path
-
-    path = app_relative_path(request).rstrip("/")
-    if path.startswith("/api/llmwiki/knowledge-bases/") and path.endswith("/weknora/frame"):
-        query_token = (request.query_params.get("access_token") or "").strip()
-        if query_token:
-            return query_token
-    return None
-
-
-def get_local_provider() -> LocalAuthProvider:
-    """Get or create the cached LocalAuthProvider singleton.
-
-    Must be called after ``init_engine_from_config()`` — the shared
-    session factory is required to construct the user repository.
-    """
-    global _cached_local_provider, _cached_repo
-    if _cached_repo is None:
-        from app.gateway.auth.repositories.sqlite import SQLiteUserRepository
-        from deerflow.persistence.engine import get_session_factory
-
-        sf = get_session_factory()
-        if sf is None:
-            raise RuntimeError("get_local_provider() called before init_engine_from_config(); cannot access users table")
-        _cached_repo = SQLiteUserRepository(sf)
-    if _cached_local_provider is None:
-        from app.gateway.auth.local_provider import LocalAuthProvider
-
-        _cached_local_provider = LocalAuthProvider(repository=_cached_repo)
-    return _cached_local_provider
-
-
-async def get_current_user_from_request(request: Request):
-    """Get the current authenticated user from request token.
-
-    Raises HTTPException 401 if not authenticated.
-    """
-    from app.gateway.auth import decode_token
-    from app.gateway.auth.errors import AuthErrorCode, AuthErrorResponse, TokenError, token_error_to_code
-
-    access_token = get_request_access_token(request)
-    if not access_token:
-        raise HTTPException(
-            status_code=401,
-            detail=AuthErrorResponse(code=AuthErrorCode.NOT_AUTHENTICATED, message="Not authenticated").model_dump(),
-        )
-
-    payload = decode_token(access_token)
-    if isinstance(payload, TokenError):
-        raise HTTPException(
-            status_code=401,
-            detail=AuthErrorResponse(code=token_error_to_code(payload), message=f"Token error: {payload.value}").model_dump(),
-        )
-
-    provider = get_local_provider()
-    user = await provider.get_user(payload.sub)
-    if user is None:
-        raise HTTPException(
-            status_code=401,
-            detail=AuthErrorResponse(code=AuthErrorCode.USER_NOT_FOUND, message="User not found").model_dump(),
-        )
-
-    # Token version mismatch → password was changed, token is stale
-    if user.token_version != payload.ver:
-        raise HTTPException(
-            status_code=401,
-            detail=AuthErrorResponse(code=AuthErrorCode.TOKEN_INVALID, message="Token revoked (password changed)").model_dump(),
-        )
-
-    return user
-
-
-async def get_optional_user_from_request(request: Request):
-    """Get optional authenticated user from request.
-
-    Returns None if not authenticated.
-    """
-    try:
-        return await get_current_user_from_request(request)
-    except HTTPException:
-        return None
-
-
-async def get_current_user(request: Request) -> str | None:
-    """Extract user_id from request cookie, or None if not authenticated.
-
-    Thin adapter that returns the string id for callers that only need
-    identification (e.g., ``feedback.py``). Full-user callers should use
-    ``get_current_user_from_request`` or ``get_optional_user_from_request``.
-    """
-    user = await get_optional_user_from_request(request)
-    return str(user.id) if user else None
+            app.state.deep_research_dispatcher = DeepResearchJobDispatcher(
+                app.state.deep_research_job_store,
+                app.state.deep_research_executor,
+            )
+            app.state.deep_research_dispatcher.start()
+        else:
+            app.state.deep_research_executor = None
+            app.state.deep_research_dispatcher = None
+
+        # Run event store (has its own factory with config-driven backend selection)
+        run_events_config = getattr(config, "run_events", None)
+        app.state.run_event_store = make_run_event_store(run_events_config)
+
+        # RunManager with store backing for persistence
+        app.state.run_manager = RunManager(store=app.state.run_store)
+        from deerflow.runtime import make_redis_concurrency_gate
+
+        app.state.concurrency_gate = make_redis_concurrency_gate(config)
+
+        try:
+            yield
+        finally:
+            from deerflow.persistence.notifications import set_notification_store
+
+            set_notification_store(None)
+            dispatcher = getattr(app.state, "roundtable_job_dispatcher", None)
+            if dispatcher is not None:
+                await dispatcher.stop()
+            dr_dispatcher = getattr(app.state, "deep_research_dispatcher", None)
+            if dr_dispatcher is not None:
+                await dr_dispatcher.stop()
+            queue = getattr(app.state, "knowledge_auto_ingest_queue", None)
+            if queue is not None:
+                from deerflow.knowledge.auto_ingest import set_auto_ingest_queue
+
+                await queue.stop()
+                set_auto_ingest_queue(None)
+            await close_engine()
+            gate = getattr(app.state, "concurrency_gate", None)
+            if gate is not None:
+                await gate.aclose()
+
+
+# ---------------------------------------------------------------------------
+# Getters – called by routers per-request
+# ---------------------------------------------------------------------------
+
+
+def _require(attr: str, label: str) -> Callable[[Request], T]:
+    """Create a FastAPI dependency that returns ``app.state.<attr>`` or 503."""
+
+    def dep(request: Request) -> T:
+        val = getattr(request.app.state, attr, None)
+        if val is None:
+            raise HTTPException(status_code=503, detail=f"{label} not available")
+        return cast(T, val)
+
+    dep.__name__ = dep.__qualname__ = f"get_{attr}"
+    return dep
+
+
+get_stream_bridge: Callable[[Request], StreamBridge] = _require("stream_bridge", "Stream bridge")
+get_run_manager: Callable[[Request], RunManager] = _require("run_manager", "Run manager")
+get_checkpointer: Callable[[Request], Checkpointer] = _require("checkpointer", "Checkpointer")
+get_run_event_store: Callable[[Request], RunEventStore] = _require("run_event_store", "Run event store")
+get_feedback_repo: Callable[[Request], FeedbackRepository] = _require("feedback_repo", "Feedback")
+get_run_store: Callable[[Request], RunStore] = _require("run_store", "Run store")
+
+
+def get_concurrency_gate(request: Request):
+    """Return the optional cross-worker concurrency gate."""
+    return getattr(request.app.state, "concurrency_gate", None)
+
+
+def get_store(request: Request):
+    """Return the global store (may be ``None`` if not configured)."""
+    return getattr(request.app.state, "store", None)
+
+
+def get_thread_store(request: Request) -> ThreadMetaStore:
+    """Return the thread metadata store (SQL or memory-backed)."""
+    val = getattr(request.app.state, "thread_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Thread metadata store not available")
+    return val
+
+
+def get_scheduled_task_store(request: Request) -> ScheduledTaskStore:
+    """Return the scheduled-task store."""
+    val = getattr(request.app.state, "scheduled_task_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Scheduled task store not available")
+    return val
+
+
+def get_llm_metrics_store(request: Request):
+    """Return the LLM metrics store, or None when not configured."""
+    return getattr(request.app.state, "llm_metrics_store", None)
+
+
+def get_concurrency_sample_store(request: Request):
+    """Return the concurrency-sample store, or None on the memory backend."""
+    return getattr(request.app.state, "concurrency_sample_store", None)
+
+
+def get_tool_metrics_store(request: Request):
+    """Return the tool-call metrics store, or None when not configured."""
+    return getattr(request.app.state, "tool_metrics_store", None)
+
+
+def get_agent_store(request: Request) -> AgentStore:
+    """Return the custom-agent store."""
+    val = getattr(request.app.state, "agent_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Agent store not available")
+    return val
+
+
+def get_skill_store(request: Request) -> SkillStore:
+    """Return the custom-skill store."""
+    val = getattr(request.app.state, "skill_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Skill store not available")
+    return val
+
+
+def get_notification_store(request: Request) -> NotificationStore:
+    """Return the in-app notification store."""
+    val = getattr(request.app.state, "notification_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Notification store not available")
+    return val
+
+
+def get_tag_store(request: Request) -> TagStore:
+    """Return the tag store."""
+    val = getattr(request.app.state, "tag_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Tag store not available")
+    return val
+
+
+def get_position_store(request: Request):
+    """Return the 岗位 (position) store."""
+    val = getattr(request.app.state, "position_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Position store not available")
+    return val
+
+
+def get_position_sync(request: Request):
+    """Return the position → subscription sync service."""
+    val = getattr(request.app.state, "position_sync", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Position sync service not available")
+    return val
+
+
+def get_thread_share_store(request: Request):
+    """Return the conversation share-link store."""
+    val = getattr(request.app.state, "thread_share_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Thread share store not available")
+    return val
+
+
+def get_roundtable_draft_share_store(request: Request):
+    """Return the roundtable-draft share-link store."""
+    val = getattr(request.app.state, "roundtable_draft_share_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Roundtable draft share store not available")
+    return val
+
+
+def get_embed_session_store(request: Request):
+    """Return the iframe embed session → thread mapping store."""
+    val = getattr(request.app.state, "embed_session_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Embed session store not available")
+    return val
+
+
+def get_ai_writing_session_store(request: Request):
+    """Return the AI writing session store, or None when not configured."""
+    return getattr(request.app.state, "ai_writing_session_store", None)
+
+
+def get_ai_writing_job_executor(request: Request):
+    """Return the AI writing background-suspend executor, or None when unavailable."""
+    return getattr(request.app.state, "ai_writing_job_executor", None)
+
+
+def get_article_type_store(request: Request):
+    """Return the article type store, or None when not configured."""
+    return getattr(request.app.state, "article_type_store", None)
+
+
+def get_research_template_store(request: Request):
+    """Return the research template store, or None when not configured."""
+    return getattr(request.app.state, "research_template_store", None)
+
+
+def get_memory_config_store(request: Request):
+    """Return the per-user memory config store."""
+    val = getattr(request.app.state, "memory_config_store", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Memory config store not available")
+    return val
+
+
+def get_knowledge_service(request: Request):
+    """Return the knowledge base service (503 when disabled / memory backend)."""
+    val = getattr(request.app.state, "knowledge_service", None)
+    if val is None:
+        raise HTTPException(status_code=503, detail="Knowledge base not available")
+    return val
+
+
+def get_run_context(request: Request) -> RunContext:
+    """Build a :class:`RunContext` from ``app.state`` singletons.
+
+    Returns a *base* context with infrastructure dependencies.
+    """
+    config = get_config(request)
+    return RunContext(
+        checkpointer=get_checkpointer(request),
+        store=get_store(request),
+        event_store=get_run_event_store(request),
+        run_events_config=getattr(config, "run_events", None),
+        thread_store=get_thread_store(request),
+        app_config=config,
+    )
+
+
+# ---------------------------------------------------------------------------
+# Auth helpers (used by authz.py and auth middleware)
+# ---------------------------------------------------------------------------
+
+# Cached singletons to avoid repeated instantiation per request
+_cached_local_provider: LocalAuthProvider | None = None
+_cached_repo: SQLiteUserRepository | None = None
+
+
+def get_request_access_token(request: Request) -> str | None:
+    """Extract access token from cookie or Authorization header.
+
+    Priority:
+    1. ``access_token`` cookie (browser session mode)
+    2. ``Authorization: Bearer <token>`` header (API client mode)
+    3. ``access_token`` query parameter for the WeKnora iframe bootstrap only
+    """
+    cookie_token = request.cookies.get("access_token")
+    if cookie_token:
+        return cookie_token
+
+    authorization = request.headers.get("authorization", "").strip()
+    if authorization:
+        scheme, _, token = authorization.partition(" ")
+        if scheme.lower() == "bearer":
+            token = token.strip()
+            if token:
+                return token
+
+    from app.gateway.proxy_path import app_relative_path
+
+    path = app_relative_path(request).rstrip("/")
+    if path.startswith("/api/llmwiki/knowledge-bases/") and path.endswith("/weknora/frame"):
+        query_token = (request.query_params.get("access_token") or "").strip()
+        if query_token:
+            return query_token
+    return None
+
+
+def get_local_provider() -> LocalAuthProvider:
+    """Get or create the cached LocalAuthProvider singleton.
+
+    Must be called after ``init_engine_from_config()`` — the shared
+    session factory is required to construct the user repository.
+    """
+    global _cached_local_provider, _cached_repo
+    if _cached_repo is None:
+        from app.gateway.auth.repositories.sqlite import SQLiteUserRepository
+        from deerflow.persistence.engine import get_session_factory
+
+        sf = get_session_factory()
+        if sf is None:
+            raise RuntimeError("get_local_provider() called before init_engine_from_config(); cannot access users table")
+        _cached_repo = SQLiteUserRepository(sf)
+    if _cached_local_provider is None:
+        from app.gateway.auth.local_provider import LocalAuthProvider
+
+        _cached_local_provider = LocalAuthProvider(repository=_cached_repo)
+    return _cached_local_provider
+
+
+async def get_current_user_from_request(request: Request):
+    """Get the current authenticated user from request token.
+
+    Raises HTTPException 401 if not authenticated.
+    """
+    from app.gateway.auth import decode_token
+    from app.gateway.auth.errors import AuthErrorCode, AuthErrorResponse, TokenError, token_error_to_code
+
+    access_token = get_request_access_token(request)
+    if not access_token:
+        raise HTTPException(
+            status_code=401,
+            detail=AuthErrorResponse(code=AuthErrorCode.NOT_AUTHENTICATED, message="Not authenticated").model_dump(),
+        )
+
+    payload = decode_token(access_token)
+    if isinstance(payload, TokenError):
+        raise HTTPException(
+            status_code=401,
+            detail=AuthErrorResponse(code=token_error_to_code(payload), message=f"Token error: {payload.value}").model_dump(),
+        )
+
+    provider = get_local_provider()
+    user = await provider.get_user(payload.sub)
+    if user is None:
+        raise HTTPException(
+            status_code=401,
+            detail=AuthErrorResponse(code=AuthErrorCode.USER_NOT_FOUND, message="User not found").model_dump(),
+        )
+
+    # Token version mismatch → password was changed, token is stale
+    if user.token_version != payload.ver:
+        raise HTTPException(
+            status_code=401,
+            detail=AuthErrorResponse(code=AuthErrorCode.TOKEN_INVALID, message="Token revoked (password changed)").model_dump(),
+        )
+
+    return user
+
+
+async def get_optional_user_from_request(request: Request):
+    """Get optional authenticated user from request.
+
+    Returns None if not authenticated.
+    """
+    try:
+        return await get_current_user_from_request(request)
+    except HTTPException:
+        return None
+
+
+async def get_current_user(request: Request) -> str | None:
+    """Extract user_id from request cookie, or None if not authenticated.
+
+    Thin adapter that returns the string id for callers that only need
+    identification (e.g., ``feedback.py``). Full-user callers should use
+    ``get_current_user_from_request`` or ``get_optional_user_from_request``.
+    """
+    user = await get_optional_user_from_request(request)
+    return str(user.id) if user else None
diff --git a/offline-backend-20260512/backend/app/gateway/routers/agents.py b/offline-backend-20260512/backend/app/gateway/routers/agents.py
index 589c2004..a60cfb1c 100644
--- a/offline-backend-20260512/backend/app/gateway/routers/agents.py
+++ b/offline-backend-20260512/backend/app/gateway/routers/agents.py
@@ -64,15 +64,20 @@ class AgentResponse(BaseModel):
         description="入库模板名 (external knowledge-ingest templateName). When set, markdown artifacts produced in this agent's chats can be published to the external knowledge-ingest API.",
     )
 
-    seat_skill_directive: bool | None = Field(
+    seat_skill_directive: bool | None = Field(
         default=None,
         description="会商席位「技能识别强化」开关 (config.yaml). True = 本智能体作为会商席位时强化技能识别 (与业务链条同名开关取 OR). None/False = 关 (默认).",
-    )
-
-    llmwiki_knowledge_base_ids: list[str] | None = Field(
-        default=None,
-        description="DeerFlow LLMWiki mapping ids this agent may offer for conversation retrieval.",
-    )
+    )
+
+    job_responsibilities: str | None = Field(
+        default=None,
+        description="岗位职责 (config.yaml). Shown above 常用问题 in the agent configuration panel. None/empty = not set.",
+    )
+
+    llmwiki_knowledge_base_ids: list[str] | None = Field(
+        default=None,
+        description="DeerFlow LLMWiki mapping ids this agent may offer for conversation retrieval.",
+    )
 
     tags: list[dict] = Field(default_factory=list, description="Tags assigned to this agent")
 
@@ -125,12 +130,13 @@ class AgentCreateRequest(BaseModel):
     tool_groups: list[str] | None = Field(default=None, description="Optional tool group whitelist persisted to config.yaml")
     common_questions: list[CommonQuestion] | None = Field(default=None, description="Recommended questions (常用问题) persisted to config.yaml. Each item is {title, prompt}; legacy strings are coerced.")
     template_name: str | None = Field(default=None, description="入库模板名 persisted to config.yaml. Enables publishing markdown artifacts to the external knowledge-ingest API.")
-    seat_skill_directive: bool | None = Field(default=None, description="会商席位「技能识别强化」开关 persisted to config.yaml. None/False = off (默认).")
-    llmwiki_knowledge_base_ids: list[str] | None = Field(
-        default=None,
-        max_length=100,
-        description="Allowed DeerFlow LLMWiki mapping ids persisted to config.yaml.",
-    )
+    seat_skill_directive: bool | None = Field(default=None, description="会商席位「技能识别强化」开关 persisted to config.yaml. None/False = off (默认).")
+    job_responsibilities: str | None = Field(default=None, description="岗位职责 persisted to config.yaml. Empty/None = not set.")
+    llmwiki_knowledge_base_ids: list[str] | None = Field(
+        default=None,
+        max_length=100,
+        description="Allowed DeerFlow LLMWiki mapping ids persisted to config.yaml.",
+    )
 
 
 class AgentUpdateRequest(BaseModel):
@@ -147,12 +153,13 @@ class AgentUpdateRequest(BaseModel):
     tool_groups: list[str] | None = Field(default=None, description="Updated tool group whitelist (config.yaml)")
     common_questions: list[CommonQuestion] | None = Field(default=None, description="Updated recommended questions (常用问题, config.yaml). Each item is {title, prompt}; legacy strings are coerced.")
     template_name: str | None = Field(default=None, description="Updated 入库模板名 (config.yaml). Empty string clears it.")
-    seat_skill_directive: bool | None = Field(default=None, description="Updated 会商席位「技能识别强化」开关 (config.yaml).")
-    llmwiki_knowledge_base_ids: list[str] | None = Field(
-        default=None,
-        max_length=100,
-        description="Updated allowed DeerFlow LLMWiki mapping ids (config.yaml).",
-    )
+    seat_skill_directive: bool | None = Field(default=None, description="Updated 会商席位「技能识别强化」开关 (config.yaml).")
+    job_responsibilities: str | None = Field(default=None, description="Updated 岗位职责 (config.yaml). Empty string clears it.")
+    llmwiki_knowledge_base_ids: list[str] | None = Field(
+        default=None,
+        max_length=100,
+        description="Updated allowed DeerFlow LLMWiki mapping ids (config.yaml).",
+    )
 
 
 class AgentSelectorResponse(BaseModel):
@@ -193,38 +200,38 @@ def _current_user_id(request: Request) -> str:
     return get_effective_user_id()
 
 
-def _is_admin_user(request: Request) -> bool:
+def _is_admin_user(request: Request) -> bool:
     """Return True when the request is authenticated as a system admin."""
     user = getattr(request.state, "user", None)
-    return getattr(user, "system_role", None) == "admin"
-
-
-async def _validate_llmwiki_bindings(
-    request: Request,
-    mapping_ids: list[str] | None,
-) -> list[str] | None:
-    """Normalize bindings and require read access for the editing actor."""
-
-    if mapping_ids is None:
-        return None
-    normalized = list(dict.fromkeys(item.strip() for item in mapping_ids if item.strip()))
-    if not normalized:
-        return []
-    store = getattr(request.app.state, "llmwiki_store", None)
-    if store is None:
-        raise HTTPException(status_code=503, detail="LLMWiki metadata store is unavailable")
-    user_id = _current_user_id(request)
-    is_admin = _is_admin_user(request)
-    for mapping_id in normalized:
-        row = await store.get_authorized(
-            mapping_id,
-            user_id,
-            write=False,
-            is_admin=is_admin,
-        )
-        if row is None:
-            raise HTTPException(status_code=422, detail=f"LLMWiki knowledge base '{mapping_id}' is unavailable")
-    return normalized
+    return getattr(user, "system_role", None) == "admin"
+
+
+async def _validate_llmwiki_bindings(
+    request: Request,
+    mapping_ids: list[str] | None,
+) -> list[str] | None:
+    """Normalize bindings and require read access for the editing actor."""
+
+    if mapping_ids is None:
+        return None
+    normalized = list(dict.fromkeys(item.strip() for item in mapping_ids if item.strip()))
+    if not normalized:
+        return []
+    store = getattr(request.app.state, "llmwiki_store", None)
+    if store is None:
+        raise HTTPException(status_code=503, detail="LLMWiki metadata store is unavailable")
+    user_id = _current_user_id(request)
+    is_admin = _is_admin_user(request)
+    for mapping_id in normalized:
+        row = await store.get_authorized(
+            mapping_id,
+            user_id,
+            write=False,
+            is_admin=is_admin,
+        )
+        if row is None:
+            raise HTTPException(status_code=422, detail=f"LLMWiki knowledge base '{mapping_id}' is unavailable")
+    return normalized
 
 
 def _email_local_part(email: str | None) -> str | None:
@@ -276,7 +283,7 @@ def _validate_storage_id(agent_id: str) -> str:
         raise HTTPException(status_code=422, detail=str(exc)) from exc
 
 
-def _write_agent_config(
+def _write_agent_config(
     agent_id: str,
     *,
     name: str,
@@ -286,9 +293,10 @@ def _write_agent_config(
     tool_groups: list[str] | None = None,
     common_questions: list[CommonQuestion] | None = None,
     template_name: str | None = None,
-    seat_skill_directive: bool | None = None,
-    llmwiki_knowledge_base_ids: list[str] | None = None,
-) -> None:
+    seat_skill_directive: bool | None = None,
+    job_responsibilities: str | None = None,
+    llmwiki_knowledge_base_ids: list[str] | None = None,
+) -> None:
     agent_dir = get_paths().agent_dir(agent_id)
     agent_dir.mkdir(parents=True, exist_ok=True)
     config_file = agent_dir / "config.yaml"
@@ -317,19 +325,30 @@ def _write_agent_config(
     if template_name is not None and template_name.strip():
         config_data["template_name"] = template_name.strip()
     # 会商席位技能识别强化开关：仅当显式设置时写入（None = 不写, 由 AgentConfig 默认为关）。
-    if seat_skill_directive is not None:
-        config_data["seat_skill_directive"] = bool(seat_skill_directive)
-    if llmwiki_knowledge_base_ids is not None:
-        config_data["llmwiki_knowledge_base_ids"] = list(
-            dict.fromkeys(item.strip() for item in llmwiki_knowledge_base_ids if item.strip())
-        )
+    if seat_skill_directive is not None:
+        config_data["seat_skill_directive"] = bool(seat_skill_directive)
+    if job_responsibilities is not None and job_responsibilities.strip():
+        config_data["job_responsibilities"] = job_responsibilities.strip()
+    if llmwiki_knowledge_base_ids is not None:
+        config_data["llmwiki_knowledge_base_ids"] = list(
+            dict.fromkeys(item.strip() for item in llmwiki_knowledge_base_ids if item.strip())
+        )
     with open(config_file, "w", encoding="utf-8") as f:
         yaml.dump(config_data, f, default_flow_style=False, allow_unicode=True)
 
 
 def _load_config_extras(agent_id: str) -> dict[str, Any]:
     """Read skills/model/tool_groups/common_questions from the agent's config.yaml, tolerant of missing files."""
-    empty = {"skills": None, "model": None, "tool_groups": None, "common_questions": None, "template_name": None, "seat_skill_directive": None, "llmwiki_knowledge_base_ids": None}
+    empty = {
+        "skills": None,
+        "model": None,
+        "tool_groups": None,
+        "common_questions": None,
+        "template_name": None,
+        "seat_skill_directive": None,
+        "job_responsibilities": None,
+        "llmwiki_knowledge_base_ids": None,
+    }
     try:
         cfg = load_agent_config(agent_id)
     except (FileNotFoundError, ValueError):
@@ -342,8 +361,9 @@ def _load_config_extras(agent_id: str) -> dict[str, Any]:
         "tool_groups": cfg.tool_groups,
         "common_questions": cfg.common_questions,
         "template_name": cfg.template_name,
-        "seat_skill_directive": cfg.seat_skill_directive,
-        "llmwiki_knowledge_base_ids": cfg.llmwiki_knowledge_base_ids,
+        "seat_skill_directive": cfg.seat_skill_directive,
+        "job_responsibilities": cfg.job_responsibilities,
+        "llmwiki_knowledge_base_ids": cfg.llmwiki_knowledge_base_ids,
     }
 
 
@@ -504,7 +524,16 @@ def _record_to_response(
             can happen lazily when the agent is actually opened.
     """
     soul: str | None = None
-    extras: dict[str, Any] = {"skills": None, "model": None, "tool_groups": None, "common_questions": None, "template_name": None, "seat_skill_directive": None, "llmwiki_knowledge_base_ids": None}
+    extras: dict[str, Any] = {
+        "skills": None,
+        "model": None,
+        "tool_groups": None,
+        "common_questions": None,
+        "template_name": None,
+        "seat_skill_directive": None,
+        "job_responsibilities": None,
+        "llmwiki_knowledge_base_ids": None,
+    }
     if include_extras:
         _ensure_agent_dir_from_record(record)
         extras = _load_config_extras(record["id"])
@@ -528,8 +557,9 @@ def _record_to_response(
         skills=extras["skills"],
         common_questions=extras.get("common_questions"),
         template_name=extras.get("template_name"),
-        seat_skill_directive=extras.get("seat_skill_directive"),
-        llmwiki_knowledge_base_ids=extras.get("llmwiki_knowledge_base_ids"),
+        seat_skill_directive=extras.get("seat_skill_directive"),
+        job_responsibilities=extras.get("job_responsibilities"),
+        llmwiki_knowledge_base_ids=extras.get("llmwiki_knowledge_base_ids"),
         featured_order=record.get("featured_order"),
     )
 
@@ -830,7 +860,7 @@ async def get_agent(agent_id: str, request: Request) -> AgentResponse:
     summary="Create Agent",
     description="Create a new user-owned agent. The display name can be Chinese and may duplicate existing names.",
 )
-async def create_agent_endpoint(request: AgentCreateRequest, http_request: Request) -> AgentResponse:
+async def create_agent_endpoint(request: AgentCreateRequest, http_request: Request) -> AgentResponse:
     _require_agents_api_enabled()
     name = _validate_display_name(request.name)
     store = get_agent_store(http_request)
@@ -839,13 +869,13 @@ async def create_agent_endpoint(request: AgentCreateRequest, http_request: Reque
         agent_id = _validate_storage_id(request.id)
         if await store.get_any(agent_id) is not None:
             raise HTTPException(status_code=409, detail=f"Agent id '{agent_id}' already exists")
-    else:
-        agent_id = uuid4().hex
-    agent_dir = get_paths().agent_dir(agent_id)
-    llmwiki_ids = await _validate_llmwiki_bindings(
-        http_request,
-        request.llmwiki_knowledge_base_ids,
-    )
+    else:
+        agent_id = uuid4().hex
+    agent_dir = get_paths().agent_dir(agent_id)
+    llmwiki_ids = await _validate_llmwiki_bindings(
+        http_request,
+        request.llmwiki_knowledge_base_ids,
+    )
 
     try:
         if agent_dir.exists():
@@ -860,8 +890,9 @@ async def create_agent_endpoint(request: AgentCreateRequest, http_request: Reque
             tool_groups=request.tool_groups,
             common_questions=request.common_questions,
             template_name=request.template_name,
-            seat_skill_directive=request.seat_skill_directive,
-            llmwiki_knowledge_base_ids=llmwiki_ids,
+            seat_skill_directive=request.seat_skill_directive,
+            job_responsibilities=request.job_responsibilities,
+            llmwiki_knowledge_base_ids=llmwiki_ids,
         )
         _write_agent_soul(agent_id, request.soul)
 
@@ -942,14 +973,15 @@ async def update_agent(agent_id: str, request: AgentUpdateRequest, http_request:
         next_tool_groups = request.tool_groups if "tool_groups" in fields_set else existing_extras["tool_groups"]
         next_common_questions = request.common_questions if "common_questions" in fields_set else existing_extras["common_questions"]
         next_template_name = request.template_name if "template_name" in fields_set else existing_extras["template_name"]
-        next_seat_skill_directive = request.seat_skill_directive if "seat_skill_directive" in fields_set else existing_extras["seat_skill_directive"]
-        next_llmwiki_ids = (
-            await _validate_llmwiki_bindings(http_request, request.llmwiki_knowledge_base_ids)
-            if "llmwiki_knowledge_base_ids" in fields_set
-            else existing_extras["llmwiki_knowledge_base_ids"]
-        )
-
-        config_dirty = bool(update_data) or any(key in fields_set for key in ("skills", "model", "tool_groups", "common_questions", "template_name", "seat_skill_directive", "llmwiki_knowledge_base_ids"))
+        next_seat_skill_directive = request.seat_skill_directive if "seat_skill_directive" in fields_set else existing_extras["seat_skill_directive"]
+        next_job_responsibilities = request.job_responsibilities if "job_responsibilities" in fields_set else existing_extras["job_responsibilities"]
+        next_llmwiki_ids = (
+            await _validate_llmwiki_bindings(http_request, request.llmwiki_knowledge_base_ids)
+            if "llmwiki_knowledge_base_ids" in fields_set
+            else existing_extras["llmwiki_knowledge_base_ids"]
+        )
+
+        config_dirty = bool(update_data) or any(key in fields_set for key in ("skills", "model", "tool_groups", "common_questions", "template_name", "seat_skill_directive", "job_responsibilities", "llmwiki_knowledge_base_ids"))
         if config_dirty:
             _write_agent_config(
                 agent_id,
@@ -960,8 +992,9 @@ async def update_agent(agent_id: str, request: AgentUpdateRequest, http_request:
                 tool_groups=next_tool_groups,
                 common_questions=next_common_questions,
                 template_name=next_template_name,
-                seat_skill_directive=next_seat_skill_directive,
-                llmwiki_knowledge_base_ids=next_llmwiki_ids,
+                seat_skill_directive=next_seat_skill_directive,
+                job_responsibilities=next_job_responsibilities,
+                llmwiki_knowledge_base_ids=next_llmwiki_ids,
             )
             # Keep the listing cache in lock-step with config.yaml, but only
             # when the cached fields were actually touched — a partial update
diff --git a/offline-backend-20260512/backend/app/gateway/routers/report_structures.py b/offline-backend-20260512/backend/app/gateway/routers/report_structures.py
new file mode 100644
index 00000000..f69bccd7
--- /dev/null
+++ b/offline-backend-20260512/backend/app/gateway/routers/report_structures.py
@@ -0,0 +1,168 @@
+"""CRUD API for deep-research report structures (报告结构 / 工作流配置).
+
+The deep-research workbench's 高级配置 → 工作流配置 dialog renders a card per
+row; picking a card applies ``content`` to the research config's
+``custom_outline``. Global / shared — every account sees the same cards.
+
+Routes (prefix ``/api/report-structures``):
+    GET    ""                list (optional ``enabled`` / ``title`` filters)
+    POST   ""                create (title / name / content; enabled forced True)
+    GET    "/{id}"           get one
+    PUT    "/{id}"           update title / name / content
+    PUT    "/{id}/enabled"   toggle 启用/停用
+    DELETE "/{id}"           delete
+
+The first list call seeds the two built-in outlines (通用分析结构 /
+递进式论证结构) when the table has never held a row — see
+``deerflow.persistence.report_structures.defaults``.
+"""
+
+from __future__ import annotations
+
+import logging
+from uuid import uuid4
+
+from fastapi import APIRouter, HTTPException, Query, Request
+from pydantic import BaseModel, Field
+
+from app.gateway.deps import get_optional_user_from_request
+from deerflow.persistence.report_structures import ReportStructureStore
+
+logger = logging.getLogger(__name__)
+router = APIRouter(prefix="/api/report-structures", tags=["report-structures"])
+
+
+# ── schemas ──────────────────────────────────────────────────────────────
+
+
+class ReportStructureResponse(BaseModel):
+    id: str
+    title: str
+    name: str
+    content: str
+    enabled: bool = True
+    created_by: str | None = None
+    created_at: str | None = None
+    updated_at: str | None = None
+
+
+class ReportStructureListResponse(BaseModel):
+    structures: list[ReportStructureResponse]
+
+
+class ReportStructureCreateRequest(BaseModel):
+    title: str = Field(..., min_length=1, max_length=256)
+    name: str = Field(default="", max_length=256)
+    content: str = Field(..., min_length=1, max_length=20_000)
+
+
+class ReportStructureUpdateRequest(BaseModel):
+    title: str | None = Field(default=None, min_length=1, max_length=256)
+    name: str | None = Field(default=None, max_length=256)
+    content: str | None = Field(default=None, min_length=1, max_length=20_000)
+
+
+class ReportStructureEnabledRequest(BaseModel):
+    enabled: bool
+
+
+# ── helpers ──────────────────────────────────────────────────────────────
+
+
+def _get_store(request: Request) -> ReportStructureStore:
+    store = getattr(request.app.state, "report_structure_store", None)
+    if store is None:
+        raise HTTPException(status_code=503, detail="Report structure store not available")
+    return store
+
+
+async def _current_username(request: Request) -> str | None:
+    """创建用户名：邮箱（本部署的账号即邮箱），无认证时为 None。"""
+    user = await get_optional_user_from_request(request)
+    if user is None:
+        return None
+    return str(user.email) or str(user.id)
+
+
+# ── routes ───────────────────────────────────────────────────────────────
+
+
+@router.get("", response_model=ReportStructureListResponse)
+async def list_report_structures(
+    request: Request,
+    enabled: bool | None = Query(default=None, description="启用/停用筛选；缺省=全部"),
+    title: str | None = Query(default=None, max_length=256, description="标题模糊筛选（大小写不敏感）"),
+) -> ReportStructureListResponse:
+    store = _get_store(request)
+    await store.ensure_defaults()
+    rows = await store.list_structures()
+    if enabled is not None:
+        rows = [r for r in rows if bool(r.get("enabled", True)) is enabled]
+    if title:
+        needle = title.strip().lower()
+        if needle:
+            rows = [r for r in rows if needle in str(r.get("title", "")).lower()]
+    return ReportStructureListResponse(structures=[ReportStructureResponse(**r) for r in rows])
+
+
+@router.post("", response_model=ReportStructureResponse, status_code=201)
+async def create_report_structure(request: Request, body: ReportStructureCreateRequest) -> ReportStructureResponse:
+    store = _get_store(request)
+    row = await store.create_structure(
+        {
+            "id": str(uuid4()),
+            "title": body.title.strip(),
+            "name": body.name.strip(),
+            "content": body.content.strip(),
+            "created_by": await _current_username(request),
+        }
+    )
+    return ReportStructureResponse(**row)
+
+
+@router.get("/{structure_id}", response_model=ReportStructureResponse)
+async def get_report_structure(request: Request, structure_id: str) -> ReportStructureResponse:
+    store = _get_store(request)
+    row = await store.get_structure(structure_id)
+    if row is None:
+        raise HTTPException(status_code=404, detail="报告结构不存在")
+    return ReportStructureResponse(**row)
+
+
+@router.put("/{structure_id}", response_model=ReportStructureResponse)
+async def update_report_structure(
+    request: Request, structure_id: str, body: ReportStructureUpdateRequest
+) -> ReportStructureResponse:
+    store = _get_store(request)
+    data: dict = {}
+    if body.title is not None:
+        data["title"] = body.title.strip()
+    if body.name is not None:
+        data["name"] = body.name.strip()
+    if body.content is not None:
+        data["content"] = body.content.strip()
+    if not data:
+        raise HTTPException(status_code=400, detail="没有需要更新的字段")
+    row = await store.update_structure(structure_id, data)
+    if row is None:
+        raise HTTPException(status_code=404, detail="报告结构不存在")
+    return ReportStructureResponse(**row)
+
+
+@router.put("/{structure_id}/enabled", response_model=ReportStructureResponse)
+async def set_report_structure_enabled(
+    request: Request, structure_id: str, body: ReportStructureEnabledRequest
+) -> ReportStructureResponse:
+    store = _get_store(request)
+    row = await store.set_enabled(structure_id, body.enabled)
+    if row is None:
+        raise HTTPException(status_code=404, detail="报告结构不存在")
+    return ReportStructureResponse(**row)
+
+
+@router.delete("/{structure_id}", status_code=204)
+async def delete_report_structure(request: Request, structure_id: str) -> None:
+    store = _get_store(request)
+    deleted = await store.delete_structure(structure_id)
+    if not deleted:
+        raise HTTPException(status_code=404, detail="报告结构不存在")
diff --git a/offline-backend-20260512/backend/app/gateway/routers/roundtable_chains.py b/offline-backend-20260512/backend/app/gateway/routers/roundtable_chains.py
index b958dc6e..86dfec4b 100644
--- a/offline-backend-20260512/backend/app/gateway/routers/roundtable_chains.py
+++ b/offline-backend-20260512/backend/app/gateway/routers/roundtable_chains.py
@@ -13,8 +13,9 @@ Routes (prefix ``/api/roundtable-chains``):
     DELETE "/{chain_id}"    delete
 
 管理员权限：``system_role == "admin"`` 的用户（``_is_admin``）可 list ``scope=all``
-（所有账号的全部链条），并可 get / 编辑 / 发布 / 删除**任意** owner 的链条；非管理员
+（所有账号的全部链条），并可 get / 编辑 / 发布 / 启停 / 删除**任意** owner 的链条；非管理员
 仍仅限本人链条（``scope=all`` 退化为 ``mine``）。编辑/删除不转移链条归属（``user_id`` 不变）。
+``enabled`` 默认 True：停用后管理列表仍可见，但选链弹窗 / 公共入口不再选用。
 """
 
 from __future__ import annotations
@@ -90,6 +91,8 @@ class ChainResponse(BaseModel):
     gather_first: bool = False
     # 发布状态：True = 公共（所有账号可在「公共」筛选查看 / 选用）。
     is_public: bool = False
+    # 启停状态：True = 启用（可选入研讨）；False = 停用。缺省 True。
+    enabled: bool = True
     # 是否当前用户所有（仅本人可编辑 / 删除 / 发布；公共筛选里别人的链条 is_mine=False）。
     is_mine: bool = True
     # 配置者（owner）的 user_id 及解析后的可读用户名（email）。owner_name 由路由层
@@ -116,6 +119,7 @@ class ChainCreateRequest(BaseModel):
     coordinator_prompt: str | None = Field(default=None, max_length=4000)
     seat_mode: str | None = Field(default=None, max_length=16)
     gather_first: bool = False
+    enabled: bool = True
 
 
 class ChainUpdateRequest(BaseModel):
@@ -131,6 +135,8 @@ class ChainUpdateRequest(BaseModel):
     gather_first: bool | None = None
     # 发布开关（True 发布为公共 / False 取消发布）；仅 owner 可改（update 本就 owner 限定）。
     is_public: bool | None = None
+    # 启停开关（True 启用 / False 停用）；仅 owner / 管理员可改。omit 不改。
+    enabled: bool | None = None
 
 
 # ── helpers ──────────────────────────────────────────────────────────────
@@ -329,6 +335,7 @@ async def create_chain(request: Request, body: ChainCreateRequest) -> ChainRespo
             "coordinator_prompt": body.coordinator_prompt,
             "seat_mode": body.seat_mode,
             "gather_first": body.gather_first,
+            "enabled": body.enabled,
         },
     )
     await _enrich_owner_names([row])
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/config/agents_config.py b/offline-backend-20260512/backend/packages/harness/deerflow/config/agents_config.py
index 17a80a11..6fef1133 100644
--- a/offline-backend-20260512/backend/packages/harness/deerflow/config/agents_config.py
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/config/agents_config.py
@@ -7,7 +7,7 @@ import threading
 from typing import Any
 
 import yaml
-from pydantic import BaseModel, field_validator, model_validator
+from pydantic import BaseModel, field_validator, model_validator
 
 from deerflow.config.paths import get_paths
 
@@ -82,7 +82,7 @@ class CommonQuestion(BaseModel):
         return value
 
 
-class AgentConfig(BaseModel):
+class AgentConfig(BaseModel):
     """Configuration for a custom agent."""
 
     id: str | None = None
@@ -108,20 +108,23 @@ class AgentConfig(BaseModel):
     # 拼进本轮任务，避免弱模型无视系统提示里的技能块。None/False = 关（默认）；True = 开。
     # 与「业务链条配置」里的同名开关取 OR：任一打开即对该席位生效。见
     # ``app/gateway/roundtable_seat_skills.py``。
-    seat_skill_directive: bool | None = None
-    # DeerFlow mapping ids this agent may offer in its new-conversation
-    # knowledge selector. WeKnora ids and credentials never enter agent files.
-    llmwiki_knowledge_base_ids: list[str] | None = None
-
-    @field_validator("llmwiki_knowledge_base_ids")
-    @classmethod
-    def _normalize_llmwiki_ids(cls, value: list[str] | None) -> list[str] | None:
-        if value is None:
-            return None
-        normalized = list(dict.fromkeys(str(item).strip() for item in value if str(item).strip()))
-        if len(normalized) > 100:
-            raise ValueError("An agent can bind at most 100 LLMWiki knowledge bases")
-        return normalized
+    seat_skill_directive: bool | None = None
+    # 岗位职责：会商智能体配置面板展示用描述。Display/UX metadata only —
+    # the runtime never reads this. None/empty = not set.
+    job_responsibilities: str | None = None
+    # DeerFlow mapping ids this agent may offer in its new-conversation
+    # knowledge selector. WeKnora ids and credentials never enter agent files.
+    llmwiki_knowledge_base_ids: list[str] | None = None
+
+    @field_validator("llmwiki_knowledge_base_ids")
+    @classmethod
+    def _normalize_llmwiki_ids(cls, value: list[str] | None) -> list[str] | None:
+        if value is None:
+            return None
+        normalized = list(dict.fromkeys(str(item).strip() for item in value if str(item).strip()))
+        if len(normalized) > 100:
+            raise ValueError("An agent can bind at most 100 LLMWiki knowledge bases")
+        return normalized
 
 
 def load_agent_config(agent_id: str | None) -> AgentConfig | None:
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/migrations/versions/20260819_02_chain_enabled.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/migrations/versions/20260819_02_chain_enabled.py
new file mode 100644
index 00000000..ba0e692e
--- /dev/null
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/migrations/versions/20260819_02_chain_enabled.py
@@ -0,0 +1,51 @@
+"""Add enabled column to roundtable_chains.
+
+Per-chain 启停 switch: True = 可在选链弹窗 / 公共入口里选用；False = 停用（管理列表
+仍可见，但不可选入研讨）。Default True → existing chains stay enabled. BOOLEAN NOT
+NULL DEFAULT 1, matching ``RoundtableChainRow.enabled``.
+
+``create_all`` adds the column for brand-new DBs; the startup ``_ensure_orm_columns_sync``
+sweep adds it to already-existing tables (sqlite/mysql/postgres). This revision keeps the
+alembic graph consistent with the rest of the chain columns and documents the change.
+
+Revision ID: 20260819_02
+Revises: 20260819_01
+Create Date: 2026-08-19
+"""
+
+from __future__ import annotations
+
+import sqlalchemy as sa
+from alembic import op
+from sqlalchemy import inspect
+
+
+revision = "20260819_02"
+down_revision = "20260819_01"
+branch_labels = None
+depends_on = None
+
+
+def _has_table(table_name: str) -> bool:
+    inspector = inspect(op.get_bind())
+    return table_name in inspector.get_table_names()
+
+
+def _has_column(table_name: str, column_name: str) -> bool:
+    if not _has_table(table_name):
+        return False
+    inspector = inspect(op.get_bind())
+    return any(col["name"] == column_name for col in inspector.get_columns(table_name))
+
+
+def upgrade() -> None:
+    if _has_table("roundtable_chains") and not _has_column("roundtable_chains", "enabled"):
+        op.add_column(
+            "roundtable_chains",
+            sa.Column("enabled", sa.Boolean(), nullable=False, server_default="1"),
+        )
+
+
+def downgrade() -> None:
+    if _has_column("roundtable_chains", "enabled"):
+        op.drop_column("roundtable_chains", "enabled")
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/__init__.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/__init__.py
new file mode 100644
index 00000000..5a3ad87f
--- /dev/null
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/__init__.py
@@ -0,0 +1,22 @@
+"""Deep-research report-structure persistence (报告结构 / 工作流配置)."""
+
+from __future__ import annotations
+
+from deerflow.persistence.report_structures.base import ReportStructureStore
+from deerflow.persistence.report_structures.memory import MemoryReportStructureStore
+from deerflow.persistence.report_structures.model import ReportStructureRow
+from deerflow.persistence.report_structures.sql import ReportStructureRepository
+
+__all__ = [
+    "MemoryReportStructureStore",
+    "ReportStructureRepository",
+    "ReportStructureRow",
+    "ReportStructureStore",
+    "make_report_structure_store",
+]
+
+
+def make_report_structure_store(session_factory):
+    if session_factory is None:
+        return MemoryReportStructureStore()
+    return ReportStructureRepository(session_factory)
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/base.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/base.py
new file mode 100644
index 00000000..2321a2e5
--- /dev/null
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/base.py
@@ -0,0 +1,64 @@
+"""Abstract report-structure persistence API (报告结构 / 工作流配置).
+
+Global / shared store: no ``user_id`` scoping — the deep-research 工作流配置
+dialog shows the same cards to every account. ``created_by`` is a display-only
+audit field (创建用户名). Rows are managed per-item (create / update /
+toggle-enabled / delete); ``ensure_defaults`` lazily seeds the two built-in
+outlines the first time the table is empty so a fresh deployment renders the
+same 通用分析结构 / 递进式论证结构 cards the frontend used to hardcode.
+"""
+
+from __future__ import annotations
+
+import logging
+from abc import ABC, abstractmethod
+from typing import Any
+
+from deerflow.persistence.report_structures.defaults import DEFAULT_REPORT_STRUCTURE_ROWS
+
+logger = logging.getLogger(__name__)
+
+
+class ReportStructureStore(ABC):
+    @abstractmethod
+    async def list_structures(self) -> list[dict[str, Any]]:
+        """Return every stored structure, most recently modified first."""
+        raise NotImplementedError
+
+    @abstractmethod
+    async def get_structure(self, structure_id: str) -> dict[str, Any] | None:
+        raise NotImplementedError
+
+    @abstractmethod
+    async def create_structure(self, data: dict[str, Any]) -> dict[str, Any]:
+        """Insert ``{id, title, name, content, created_by}`` (enabled forced True)."""
+        raise NotImplementedError
+
+    @abstractmethod
+    async def update_structure(self, structure_id: str, data: dict[str, Any]) -> dict[str, Any] | None:
+        """Update any of ``title`` / ``name`` / ``content``. None when missing."""
+        raise NotImplementedError
+
+    @abstractmethod
+    async def set_enabled(self, structure_id: str, enabled: bool) -> dict[str, Any] | None:
+        """Flip 启用/停用 (also bumps updated_at). None when missing."""
+        raise NotImplementedError
+
+    @abstractmethod
+    async def delete_structure(self, structure_id: str) -> bool:
+        raise NotImplementedError
+
+    @abstractmethod
+    async def count(self) -> int:
+        raise NotImplementedError
+
+    async def ensure_defaults(self) -> None:
+        """Seed the built-in outlines when the table has never held a row.
+
+        Concurrent first reads can race; seed ids are fixed so a losing insert
+        fails on the primary key and is simply skipped.
+        """
+        if await self.count() > 0:
+            return
+        for row in DEFAULT_REPORT_STRUCTURE_ROWS:
+            await self.create_structure(dict(row))
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/defaults.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/defaults.py
new file mode 100644
index 00000000..be0c3db7
--- /dev/null
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/defaults.py
@@ -0,0 +1,269 @@
+"""深度研究「报告结构」默认种子（新机器部署开箱即有的内置结构库）。
+
+前两套（通用分析结构 / 递进式论证结构）逐字提取自前端
+``frontend-web/src/pages/deep-research/ReportConfigCard.tsx`` 的
+``DEFAULT_REPORT_OUTLINE`` / ``PROGRESSIVE_REPORT_OUTLINE``；其余三套为本仓库
+补充的常用研究结构（快速简报 / 事件分析 / 对象调查）。
+表为空（全新部署、从未写过数据）时 ``ensure_defaults`` 会把全部内置结构插入
+``report_structures``；已有数据的库不会追加。若调优了某套结构，请同步更新
+本文件，否则新机器补建出来的还是旧版。
+"""
+
+from __future__ import annotations
+
+GENERAL_OUTLINE = """# 报告标题
+
+# 引言
+用 1–2 段说明研究对象、提出本议题的背景、研究范围及资料边界；简要交代全文的分析路径，不单独拆分小节。
+
+# 一、研究对象与事实基础
+## 1.1 概念、范围与背景
+界定研究对象、关键概念、时间和空间范围；说明读者理解问题所需的背景。
+## 1.2 已有事实与证据
+按时间或逻辑顺序梳理已知事实、数据、案例或材料发现；优先使用可核验的来源。
+## 1.3 当前状态与主要特征
+归纳研究对象目前呈现的状态、关键变化、共性与差异；没有数据支撑时明确说明。
+
+# 二、问题形成与作用机制
+## 2.1 直接原因与触发条件
+分析近期事件、条件变化或关键节点如何推动问题出现或演变。
+## 2.2 深层影响因素
+从制度、技术、资源、行为、环境或历史条件中选择适用维度，分析长期影响。
+## 2.3 因果关系与相互作用
+说明不同因素如何共同作用；区分来源已证实的事实、分析判断和仍待验证的推断。
+
+# 三、影响、问题与不同观点
+## 3.1 对研究对象的直接影响
+从结果、效率、质量、风险、成本、体验或其他适用维度，分析当前影响。
+## 3.2 对相关主体与外部环境的影响
+说明影响如何传导到相关群体、组织、地区、系统或后续事项；不适用的维度可省略。
+## 3.3 主要争议与解释差异
+并列材料支持的不同观点、解释或结论，并说明各自的证据基础与局限。
+
+# 四、比较、趋势与不确定性
+## 4.1 可比案例或横向比较
+在材料允许时选择历史案例、不同地区、不同方案或不同阶段进行比较，说明异同与启示。
+## 4.2 发展趋势与关键变量
+识别后续可能变化、决定性变量、观察指标和潜在转折点；不把预测写成既成事实。
+## 4.3 风险情景与不确定性
+说明不同情景的触发条件、可能影响和证据不足之处；对没有依据的推测明确标注。
+
+# 五、结论、建议与后续关注
+## 5.1 核心结论
+用编号或分点方式收束全文；每条结论都应能回溯到正文中的事实和分析。
+## 5.2 可行建议或应对方向
+仅在研究材料支持时提出建议；说明适用对象、前提条件、优先级和预期作用。
+## 5.3 资料局限与后续关注
+列出证据缺口、数据口径差异、尚不能下结论的问题，以及值得持续跟踪的事项。
+
+# 参考资料
+仅列出本报告实际使用的来源，并与正文引用保持一致。
+
+> 写作要求：引言保持简洁；每个二级小节按材料充分程度写 2–5 段完整分析，每段围绕一个明确观点展开，采用“事实/证据 → 分析 → 影响或小结”的顺序。章节可根据研究主题合并、增删或调整，但不要为了套模板而编造不适用的主体、数据、案例或引用。"""
+
+PROGRESSIVE_OUTLINE = """# 报告标题
+
+# 引言
+用 1–2 段说明研究对象、核心问题、研究范围和资料边界；简要说明本文将按“事实基线 → 成因机制 → 影响判断 → 应对方向 → 结论”的递进逻辑展开。
+
+# 一、事实基线：问题是什么
+## 1.1 概念界定与研究范围
+明确关键概念、时间/对象边界和判断口径，避免后续讨论混用概念。
+## 1.2 关键事实、数据与证据
+按时间或逻辑顺序列出已经发生的事实、可核验数据、案例和来源。
+## 1.3 基线判断与待解问题
+基于前两节归纳当前状态、主要矛盾和需要解释的核心问题。
+
+> 本章产出：后续分析必须以本章已经确认的事实和问题为基础，不得在后文补写没有来源支撑的新事实。
+
+# 二、形成机制：为什么会这样
+## 2.1 直接原因与触发条件
+从第一章识别的问题出发，分析近期事件、条件变化或关键节点如何导致当前结果。
+## 2.2 深层条件与关键约束
+结合第一章的证据，说明制度、技术、资源、行为、环境或历史条件等长期影响。
+## 2.3 因果链条与关键变量
+把前两节串成清晰的“条件 → 机制 → 结果”链条，标出决定性变量及仍待验证的环节。
+
+> 本章产出：第三章只能沿用本章已说明的机制评估影响，不能把相关性直接写成因果关系。
+
+# 三、结果与影响：会带来什么
+## 3.1 直接结果与短期影响
+依据第二章的因果链条，分析对研究对象最直接、最先出现的影响。
+## 3.2 传导效应与中长期影响
+继续说明影响如何传导至相关主体、外部环境或后续事项；不适用的维度可省略。
+## 3.3 权衡、分歧与边界
+比较不同影响之间的取舍、材料支持的不同观点，以及结论适用的边界。
+
+> 本章产出：第四章的趋势判断、情景和建议应以本章已评估的影响、风险和权衡为依据。
+
+# 四、趋势与应对：接下来怎么办
+## 4.1 未来走势与观察指标
+从第三章的影响出发，识别可能变化、关键观察指标和潜在转折点；不把预测写成既成事实。
+## 4.2 可选方案与适用条件
+针对第三章发现的问题和风险，提出可行的应对方向、前提条件与优先级。
+## 4.3 风险情景与验证重点
+说明不同情景的触发条件、可能后果，以及下一步需要补充验证的数据或材料。
+
+> 本章产出：第五章只综合前四章已经建立的事实、机制、影响和建议，不新增独立论点。
+
+# 五、综合结论与后续关注
+## 5.1 核心结论
+按“事实基线 → 原因机制 → 主要影响 → 应对方向”逐项收束，每条结论都可回溯到前文证据。
+## 5.2 局限与后续研究
+列出资料缺口、数据口径差异、尚不能下结论的问题和后续跟踪清单。
+
+# 参考资料
+仅列出本报告实际使用的来源，并与正文引用保持一致。
+
+> 写作要求：除引言外，每个二级小节写 2–5 段完整分析。后章必须明确承接前章的结论；若材料无法支持递进关系，应说明资料限制而非编造衔接。"""
+
+BRIEF_OUTLINE = """# 报告标题
+
+# 核心结论
+用 3–5 条要点直接给出最重要的结论与判断，每条一两句话并可回溯到后文依据；先说结论，再列支撑。
+
+# 关键事实
+## 事实清单
+按重要性列出已核验的关键事实、数据与出处；不确定的信息单独注明置信度。
+## 时间与背景
+用最短的篇幅交代理解结论所需的事件顺序与背景，不展开细节。
+
+# 风险与不确定性
+列出可能推翻或削弱上述结论的因素、证据缺口和正在变化中的变量。
+
+# 建议行动
+按优先级给出可执行的建议或下一步关注点，说明适用前提与预期作用；材料不足时明确说明证据尚不支持。
+
+# 参考资料
+仅列出本简报实际使用的来源，并与正文引用保持一致。
+
+> 写作要求：全文控制在较短篇幅（建议 800–1500 字），信息密度优先；不写套话与背景铺陈，所有论断要么有来源支撑，要么明确标注为分析判断。"""
+
+
+EVENT_OUTLINE = """# 报告标题
+
+# 引言
+用 1–2 段说明事件的性质、发生时间地点、当前状态与本报告要回答的核心问题。
+
+# 一、事件概述
+## 1.1 事件是什么
+客观描述发生了什么：主体、行为、规模、直接结果；只写已核验事实。
+## 1.2 各方说法
+并列当事方、官方与媒体的不同表述，注明各自的信息来源与立场。
+
+# 二、时间线还原
+## 2.1 起因与前情
+梳理事件发生的背景、前置条件与诱发节点。
+## 2.2 关键节点
+按时间顺序还原事件演进的关键节点，每个节点注明时间与来源；存在争议的时间点单独标注。
+## 2.3 当前状态
+说明事件目前的进展、处置情况与未决事项。
+
+# 三、原因与责任分析
+## 3.1 直接原因
+分析触发事件的直接因素与机制。
+## 3.2 深层原因
+从制度、管理、技术、环境等维度分析更深层的促成条件。
+## 3.3 责任与归因
+基于事实分析各方责任，区分已认定的结论与仍在调查的部分。
+
+# 四、影响与反应
+## 4.1 直接影响
+评估事件对当事方与相关群体已造成的影响。
+## 4.2 各方反应与后续动作
+梳理官方处置、当事方回应、舆论反应与衍生事件。
+## 4.3 潜在延伸影响
+分析可能的中长期影响与传导路径，注明判断依据。
+
+# 五、结论与后续关注
+## 5.1 核心结论
+收束事件性质、原因与影响的核心判断。
+## 5.2 待验证问题与观察点
+列出尚未证实的关键问题与值得持续跟踪的信号。
+
+# 参考资料
+仅列出本报告实际使用的来源，并与正文引用保持一致。
+
+> 写作要求：严格区分“已证实事实—各方声称—分析推断”三个层次并分别标注；时间线每个节点须有来源；对争议点呈现多方说法，不单方采信。"""
+
+PROFILE_OUTLINE = """# 报告标题
+
+# 引言
+说明调查对象的界定范围（个人、组织或产品）、研究目的与资料边界（公开信息的时间范围与覆盖面）。
+
+# 一、基本情况
+## 1.1 基本信息与背景
+梳理对象的公认基本信息与来历，注明信息时效。
+## 1.2 发展沿革
+按时间梳理关键成长节点、转折与现状。
+
+# 二、能力与表现
+## 2.1 核心能力或竞争力
+分析对象的核心能力、资源禀赋或产品力，给出可观察的证据。
+## 2.2 代表性成果与表现
+列举代表性成果、业绩或事件表现，并评估其含金量与可持续性。
+## 2.3 短板与争议
+并列材料中的短板、批评与争议点，说明各方依据。
+
+# 三、关系网络与生态
+## 3.1 关联方与合作关系
+梳理重要的合作关系、关联方与利益结构。
+## 3.2 立场与倾向
+基于言行记录分析对象的立场、风格与决策倾向。
+
+# 四、风险评估
+## 4.1 已知风险
+汇总已暴露的风险点、负面记录与可靠性问题。
+## 4.2 潜在风险与信息缺口
+指出资料不足导致的盲区与需要进一步核实的问题。
+
+# 五、综合评估
+## 5.1 画像结论
+用几句话给出对象的综合画像与可信度评估。
+## 5.2 交往或合作建议
+给出与对象打交道的注意事项与建议，说明前提。
+
+# 参考资料
+仅列出本报告实际使用的来源，并与正文引用保持一致。
+
+> 写作要求：只依据公开可核验信息，猜测与推断必须明确标注；负面信息须注明来源与时间，区分“事实—评价—传闻”；信息不足时如实说明，不为了完整而编造。"""
+
+
+# 固定 id：并发首查时重复插入会撞主键而被吞掉，不会产生重复行。
+DEFAULT_REPORT_STRUCTURE_ROWS: list[dict] = [
+    {
+        "id": "seed-report-outline-general",
+        "title": "通用分析结构",
+        "name": "通用分析结构",
+        "content": GENERAL_OUTLINE,
+        "created_by": "系统内置",
+    },
+    {
+        "id": "seed-report-outline-progressive",
+        "title": "递进式论证结构",
+        "name": "递进式论证结构",
+        "content": PROGRESSIVE_OUTLINE,
+        "created_by": "系统内置",
+    },
+    {
+        "id": "seed-report-outline-brief",
+        "title": "快速简报结构",
+        "name": "快速简报结构",
+        "content": BRIEF_OUTLINE,
+        "created_by": "系统内置",
+    },
+    {
+        "id": "seed-report-outline-event",
+        "title": "事件分析结构",
+        "name": "事件分析结构",
+        "content": EVENT_OUTLINE,
+        "created_by": "系统内置",
+    },
+    {
+        "id": "seed-report-outline-profile",
+        "title": "对象调查结构",
+        "name": "对象调查结构",
+        "content": PROFILE_OUTLINE,
+        "created_by": "系统内置",
+    },
+]
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/memory.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/memory.py
new file mode 100644
index 00000000..ee383721
--- /dev/null
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/memory.py
@@ -0,0 +1,66 @@
+"""In-memory report-structure store (auth-less tests / no-DB fallback)."""
+
+from __future__ import annotations
+
+from datetime import UTC, datetime
+from typing import Any
+
+from deerflow.persistence.report_structures.base import ReportStructureStore
+
+
+def _now_iso() -> str:
+    return datetime.now(UTC).isoformat()
+
+
+def _sort_rows(rows: list[dict[str, Any]]) -> list[dict[str, Any]]:
+    return sorted(rows, key=lambda r: (r.get("updated_at") or "", r.get("created_at") or ""), reverse=True)
+
+
+class MemoryReportStructureStore(ReportStructureStore):
+    def __init__(self) -> None:
+        self._rows: dict[str, dict[str, Any]] = {}
+
+    async def list_structures(self) -> list[dict[str, Any]]:
+        return _sort_rows(list(self._rows.values()))
+
+    async def get_structure(self, structure_id: str) -> dict[str, Any] | None:
+        row = self._rows.get(structure_id)
+        return dict(row) if row else None
+
+    async def create_structure(self, data: dict[str, Any]) -> dict[str, Any]:
+        row = {
+            "id": str(data["id"]),
+            "title": str(data.get("title") or ""),
+            "name": str(data.get("name") or ""),
+            "content": str(data.get("content") or ""),
+            "enabled": True,
+            "created_by": data.get("created_by"),
+            "created_at": _now_iso(),
+            "updated_at": _now_iso(),
+        }
+        self._rows[row["id"]] = row
+        return dict(row)
+
+    async def update_structure(self, structure_id: str, data: dict[str, Any]) -> dict[str, Any] | None:
+        row = self._rows.get(structure_id)
+        if row is None:
+            return None
+        for key in ("title", "name", "content"):
+            if key in data:
+                row[key] = str(data[key])
+        row["updated_at"] = _now_iso()
+        return dict(row)
+
+    async def set_enabled(self, structure_id: str, enabled: bool) -> dict[str, Any] | None:
+        row = self._rows.get(structure_id)
+        if row is None:
+            return None
+        row["enabled"] = bool(enabled)
+        row["updated_at"] = _now_iso()
+        return dict(row)
+
+    async def delete_structure(self, structure_id: str) -> bool:
+        return self._rows.pop(structure_id, None) is not None
+
+    async def count(self) -> int:
+        return len(self._rows)
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/model.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/model.py
new file mode 100644
index 00000000..ba24105b
--- /dev/null
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/model.py
@@ -0,0 +1,45 @@
+"""ORM model for deep-research report structures (报告结构 / 工作流配置).
+
+Global / shared — every account sees the same structure cards in the deep
+research workbench's 工作流配置 dialog. Each row is one reusable report
+structure the user can pick to fill ``custom_outline``:
+
+- ``title`` — 标题, the card's primary label and the list's 标题筛选 target.
+- ``name`` — 名称, a secondary display label.
+- ``content`` — 接口内容, the Markdown outline text applied to the research
+  config's ``custom_outline`` when the card is selected.
+- ``enabled`` — 启用/停用; creation always defaults to True (the caller cannot
+  choose it) and a dedicated toggle flips it later.
+- ``created_by`` — 创建用户名 (audit-only display field).
+
+Fresh deployments get the two built-in outlines (通用分析结构 / 递进式论证结构)
+seeded by ``ensure_defaults`` — see ``defaults.py``.
+"""
+
+from __future__ import annotations
+
+from datetime import UTC, datetime
+
+from sqlalchemy import Boolean, String, Text
+from sqlalchemy.orm import Mapped, mapped_column
+
+from deerflow.persistence.base import Base
+from deerflow.persistence.types import BeijingDateTime
+
+
+class ReportStructureRow(Base):
+    __tablename__ = "report_structures"
+
+    id: Mapped[str] = mapped_column(String(64), primary_key=True)
+    title: Mapped[str] = mapped_column(String(256), nullable=False)
+    name: Mapped[str] = mapped_column(String(256), default="", nullable=False)
+    content: Mapped[str] = mapped_column(Text, nullable=False)
+    # ``server_default`` 必须显式声明：否则在已存在该表但缺列的旧库上，启动期
+    # ``_ensure_orm_columns_sync`` 会跳过「NOT NULL 且无 server_default」的列（见
+    # engine.py），该列永远补不上 → 查询报 Unknown column → 500。
+    enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True, server_default="1")
+    created_by: Mapped[str | None] = mapped_column(String(128), nullable=True)
+    created_at: Mapped[datetime] = mapped_column(BeijingDateTime(), nullable=False, default=lambda: datetime.now(UTC))
+    updated_at: Mapped[datetime] = mapped_column(
+        BeijingDateTime(), nullable=False, default=lambda: datetime.now(UTC), onupdate=lambda: datetime.now(UTC)
+    )
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/sql.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/sql.py
new file mode 100644
index 00000000..18cb3872
--- /dev/null
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/report_structures/sql.py
@@ -0,0 +1,108 @@
+"""SQLAlchemy-backed report-structure store (global / shared)."""
+
+from __future__ import annotations
+
+import logging
+from datetime import UTC, datetime
+from typing import Any
+
+from sqlalchemy import delete, func, select
+from sqlalchemy.exc import IntegrityError
+from sqlalchemy.ext.asyncio import AsyncSession, async_sessionmaker
+
+from deerflow.persistence.report_structures.base import ReportStructureStore
+from deerflow.persistence.report_structures.model import ReportStructureRow
+
+logger = logging.getLogger(__name__)
+
+
+def _to_dict(row: ReportStructureRow) -> dict[str, Any]:
+    data = row.to_dict()
+    for key in ("created_at", "updated_at"):
+        value = data.get(key)
+        if isinstance(value, datetime):
+            data[key] = value.isoformat()
+    return data
+
+
+class ReportStructureRepository(ReportStructureStore):
+    def __init__(self, session_factory: async_sessionmaker[AsyncSession]) -> None:
+        self._sf = session_factory
+
+    async def list_structures(self) -> list[dict[str, Any]]:
+        stmt = select(ReportStructureRow).order_by(
+            ReportStructureRow.updated_at.desc(), ReportStructureRow.created_at.asc()
+        )
+        async with self._sf() as session:
+            result = await session.execute(stmt)
+            return [_to_dict(row) for row in result.scalars()]
+
+    async def get_structure(self, structure_id: str) -> dict[str, Any] | None:
+        async with self._sf() as session:
+            row = await session.get(ReportStructureRow, structure_id)
+            return _to_dict(row) if row is not None else None
+
+    async def create_structure(self, data: dict[str, Any]) -> dict[str, Any]:
+        now = datetime.now(UTC)
+        row = ReportStructureRow(
+            id=str(data["id"]),
+            title=str(data.get("title") or ""),
+            name=str(data.get("name") or ""),
+            content=str(data.get("content") or ""),
+            # 创建永远默认启用（前端不提供该项填写）。
+            enabled=True,
+            created_by=data.get("created_by"),
+            created_at=now,
+            updated_at=now,
+        )
+        async with self._sf() as session:
+            session.add(row)
+            await session.commit()
+            await session.refresh(row)
+            return _to_dict(row)
+
+    async def update_structure(self, structure_id: str, data: dict[str, Any]) -> dict[str, Any] | None:
+        async with self._sf() as session:
+            row = await session.get(ReportStructureRow, structure_id)
+            if row is None:
+                return None
+            for key in ("title", "name", "content"):
+                if key in data:
+                    setattr(row, key, str(data[key]))
+            row.updated_at = datetime.now(UTC)
+            await session.commit()
+            await session.refresh(row)
+            return _to_dict(row)
+
+    async def set_enabled(self, structure_id: str, enabled: bool) -> dict[str, Any] | None:
+        async with self._sf() as session:
+            row = await session.get(ReportStructureRow, structure_id)
+            if row is None:
+                return None
+            row.enabled = bool(enabled)
+            row.updated_at = datetime.now(UTC)
+            await session.commit()
+            await session.refresh(row)
+            return _to_dict(row)
+
+    async def delete_structure(self, structure_id: str) -> bool:
+        async with self._sf() as session:
+            result = await session.execute(delete(ReportStructureRow).where(ReportStructureRow.id == structure_id))
+            await session.commit()
+            return bool(result.rowcount)
+
+    async def count(self) -> int:
+        async with self._sf() as session:
+            return int(await session.scalar(select(func.count()).select_from(ReportStructureRow)) or 0)
+
+    async def ensure_defaults(self) -> None:
+        """Seed when empty; a concurrent first-read loses on the fixed PK silently."""
+        if await self.count() > 0:
+            return
+        from deerflow.persistence.report_structures.defaults import DEFAULT_REPORT_STRUCTURE_ROWS
+
+        for row in DEFAULT_REPORT_STRUCTURE_ROWS:
+            try:
+                await self.create_structure(dict(row))
+            except IntegrityError:
+                logger.debug("report-structure default '%s' already inserted by a concurrent request", row["id"])
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/model.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/model.py
index bba96994..8aeed00f 100644
--- a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/model.py
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/model.py
@@ -63,6 +63,12 @@ class RoundtableChainRow(Base):
     is_public: Mapped[bool] = mapped_column(
         Boolean, nullable=False, default=False, server_default="0", index=True
     )
+    # 启停开关：True = 启用（可在选链弹窗 / 公共入口里选用）；False = 停用（管理列表仍可见，
+    # 但不可选入研讨）。默认启用，存量链条补列后也为 True。``server_default`` 必须显式声明，
+    # 否则既有表上启动期 ``_ensure_orm_columns_sync`` 会跳过「NOT NULL 且无 server_default」的列。
+    enabled: Mapped[bool] = mapped_column(
+        Boolean, nullable=False, default=True, server_default="1"
+    )
     created_at: Mapped[datetime] = mapped_column(
         BeijingDateTime(), nullable=False, default=lambda: datetime.now(UTC)
     )
diff --git a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/sql.py b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/sql.py
index 756eca86..fac6c8cf 100644
--- a/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/sql.py
+++ b/offline-backend-20260512/backend/packages/harness/deerflow/persistence/roundtable_chains/sql.py
@@ -62,6 +62,8 @@ class RoundtableChainRepository:
             "gather_first": bool(getattr(row, "gather_first", False)),
             # 发布状态 + 是否本人所有（前端据此显示「公共」徽标、决定能否编辑 / 发布）。
             "is_public": bool(getattr(row, "is_public", False)),
+            # 启停状态：缺省 True（存量行 / 补列前视为启用）。
+            "enabled": bool(getattr(row, "enabled", True)),
             "is_mine": (current_user_id is None) or (row.user_id == current_user_id),
             # 配置者（owner）的 user_id；路由层据此解析成可读用户名（owner_name）。
             "owner_id": row.user_id,
@@ -122,6 +124,7 @@ class RoundtableChainRepository:
             seat_mode=(data.get("seat_mode") or None),
             gather_first=bool(data.get("gather_first") or False),
             is_public=bool(data.get("is_public") or False),
+            enabled=True if data.get("enabled") is None else bool(data.get("enabled")),
             created_at=now,
             updated_at=now,
         )
@@ -145,6 +148,7 @@ class RoundtableChainRepository:
         seat_mode=_UNSET,
         gather_first=_UNSET,
         is_public=_UNSET,
+        enabled=_UNSET,
         is_admin: bool = False,
     ) -> dict[str, Any] | None:
         async with self._sf() as session:
@@ -170,6 +174,8 @@ class RoundtableChainRepository:
                 row.gather_first = bool(gather_first)
             if is_public is not _UNSET and is_public is not None:
                 row.is_public = bool(is_public)
+            if enabled is not _UNSET and enabled is not None:
+                row.enabled = bool(enabled)
             row.updated_at = datetime.now(UTC)
             await session.commit()
             await session.refresh(row)
diff --git a/offline-backend-20260512/backend/scripts/sql/column_additions_mysql.sql b/offline-backend-20260512/backend/scripts/sql/column_additions_mysql.sql
index 34926d2e..af410e69 100644
--- a/offline-backend-20260512/backend/scripts/sql/column_additions_mysql.sql
+++ b/offline-backend-20260512/backend/scripts/sql/column_additions_mysql.sql
@@ -94,6 +94,11 @@ CALL _add_column_if_missing('roundtable_chains', 'seat_mode', 'VARCHAR(16) NULL
 --               ORM: RoundtableChainRow.gather_first = Boolean, NOT NULL, server_default '0'。
 CALL _add_column_if_missing('roundtable_chains', 'gather_first', 'TINYINT(1) NOT NULL DEFAULT 0');
 
+-- enabled: 业务链条启停开关,对应 migration 20260819_02。True=启用(可选入研讨);
+--          False=停用(管理列表仍可见)。默认启用,存量行补列后也为 True。
+--          ORM: RoundtableChainRow.enabled = Boolean, NOT NULL, server_default '1'。
+CALL _add_column_if_missing('roundtable_chains', 'enabled', 'TINYINT(1) NOT NULL DEFAULT 1');
+
 -- ---- 岗位 (positions) 订阅系统,对应 migration 20260616_01 -------------------
 -- positions / position_tag_bindings 是【新表】,由 create_all 自动建,无需补列。
 -- 以下是给【已存在的表】补的列:
@@ -212,5 +217,6 @@ SELECT table_name, column_name
         OR (table_name = 'roundtable_jobs' AND column_name = 'task_id')
         OR (table_name = 'position_roundtable_sessions' AND column_name = 'task_scoped')
         OR (table_name = 'task_buttons' AND column_name IN ('append_auth', 'auth_token_param', 'auth_name_param'))
-        OR (table_name = 'compact_menu_overrides' AND column_name = 'menu_type'))
+        OR (table_name = 'compact_menu_overrides' AND column_name = 'menu_type')
+        OR (table_name = 'roundtable_chains' AND column_name = 'enabled'))
  ORDER BY table_name, column_name;
diff --git a/offline-backend-20260512/backend/tests/test_agents_backend.py b/offline-backend-20260512/backend/tests/test_agents_backend.py
index f87df609..b9a033fc 100644
--- a/offline-backend-20260512/backend/tests/test_agents_backend.py
+++ b/offline-backend-20260512/backend/tests/test_agents_backend.py
@@ -26,7 +26,7 @@ from deerflow.persistence.tags.memory import MemoryTagStore
 from deerflow.services.position_sync import PositionSyncService
 
 
-@pytest.fixture
+@pytest.fixture
 def isolated_paths(tmp_path, monkeypatch):
     """Redirect get_paths() at a fresh temp directory for the duration of one test."""
     monkeypatch.setenv("DEER_FLOW_HOME", str(tmp_path))
@@ -119,20 +119,20 @@ async def _default_positions() -> MemoryPositionStore:
         }
     )
     return positions
-
-
-@pytest.mark.asyncio
+
+
+@pytest.mark.asyncio
 async def test_agent_store_allows_duplicate_chinese_names_and_filters_visibility() -> None:
     store = MemoryAgentStore()
     chinese_name = "\u4e2d\u6587\u52a9\u624b"
 
-    await store.ensure_builtin({"id": "builtin-agent", "name": chinese_name, "description": "built in"})
-    await store.create({"id": "private-a", "name": chinese_name, "description": "private", "user_id": "user-a"})
-    await store.create({"id": "published-b", "name": chinese_name, "description": "published", "user_id": "user-b", "published": True})
-
-    user_a_ids = {agent["id"] for agent in await store.list_visible("user-a")}
-    assert user_a_ids == {"builtin-agent", "private-a", "published-b"}
-
+    await store.ensure_builtin({"id": "builtin-agent", "name": chinese_name, "description": "built in"})
+    await store.create({"id": "private-a", "name": chinese_name, "description": "private", "user_id": "user-a"})
+    await store.create({"id": "published-b", "name": chinese_name, "description": "published", "user_id": "user-b", "published": True})
+
+    user_a_ids = {agent["id"] for agent in await store.list_visible("user-a")}
+    assert user_a_ids == {"builtin-agent", "private-a", "published-b"}
+
     user_c_ids = {agent["id"] for agent in await store.list_visible("user-c")}
     assert user_c_ids == {"builtin-agent", "published-b"}
 
@@ -271,254 +271,297 @@ async def test_manual_favorite_upgrades_existing_auto_builtin_favorite(sql_sessi
 async def test_agent_store_only_owner_can_update_or_delete() -> None:
     store = MemoryAgentStore()
 
-    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
-    await store.create({"id": "private-a", "name": "Owned", "user_id": "user-a"})
-
-    assert await store.update("builtin-agent", "user-a", {"name": "Nope"}) is None
-    assert await store.update("private-a", "user-b", {"name": "Nope"}) is None
-
-    updated = await store.update("private-a", "user-a", {"name": "Renamed"})
-    assert updated is not None
-    assert updated["name"] == "Renamed"
-
-    assert await store.delete("private-a", "user-b") is False
-    assert await store.delete("private-a", "user-a") is True
-
-
-@pytest.mark.asyncio
-async def test_delete_builtin_only_targets_builtins() -> None:
-    store = MemoryAgentStore()
-
-    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
-    await store.create({"id": "owned-agent", "name": "Owned", "user_id": "user-a"})
-
-    # delete_builtin refuses to touch a user-owned agent
-    assert await store.delete_builtin("owned-agent") is False
-    assert await store.get_any("owned-agent") is not None
-
-    # delete_builtin succeeds on a built-in
-    assert await store.delete_builtin("builtin-agent") is True
-    assert await store.get_any("builtin-agent") is None
-
-    # And no-ops on a missing id
-    assert await store.delete_builtin("does-not-exist") is False
-
-
-@pytest.mark.asyncio
-async def test_update_builtin_only_targets_builtins() -> None:
-    store = MemoryAgentStore()
-
-    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
-    await store.create({"id": "owned-agent", "name": "Owned", "user_id": "user-a"})
-
-    # update_builtin succeeds on a built-in agent
-    updated = await store.update_builtin("builtin-agent", {"name": "Renamed by admin"})
-    assert updated is not None
-    assert updated["name"] == "Renamed by admin"
-
-    # update_builtin refuses to touch a user-owned agent (returns None)
-    assert await store.update_builtin("owned-agent", {"name": "Hijacked"}) is None
-
-    # And on a missing agent
-    assert await store.update_builtin("does-not-exist", {"name": "Ghost"}) is None
-
-
-@pytest.mark.asyncio
-async def test_get_editable_or_raise_admin_can_edit_builtin() -> None:
-    """Route helper: admin gets edit access to any agent; non-admin still 403."""
-    from fastapi import HTTPException
-
-    from app.gateway.routers.agents import _get_editable_or_raise
-
-    store = MemoryAgentStore()
-    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
-    await store.create({"id": "owned-agent", "name": "Owned", "user_id": "user-a"})
-
-    # Owner edits their own agent → returns ('owner', record)
-    record, mode = await _get_editable_or_raise(
-        store, "owned-agent", "user-a", is_admin=False
-    )
-    assert mode == "owner" and record["id"] == "owned-agent"
-
-    # Admin edits a built-in agent → returns ('admin', record)
-    record, mode = await _get_editable_or_raise(
-        store, "builtin-agent", "user-a", is_admin=True
-    )
-    assert mode == "admin" and record["id"] == "builtin-agent"
-
-    # Non-admin tries to edit a built-in → 403
-    with pytest.raises(HTTPException) as exc:
-        await _get_editable_or_raise(store, "builtin-agent", "user-a", is_admin=False)
-    assert exc.value.status_code == 403
-
-    # Admin edits someone else's custom agent → returns ('admin', record).
-    # Behavior changed: admins manage the full agent catalog (built-in + any
-    # user's agents), not just built-ins. Non-admins are still owner-only.
-    await store.create(
-        {"id": "published-other", "name": "Pub", "user_id": "user-x", "published": True}
-    )
-    record, mode = await _get_editable_or_raise(
-        store, "published-other", "user-a", is_admin=True
-    )
-    assert mode == "admin" and record["id"] == "published-other"
-
-
-def test_build_run_config_injects_custom_assistant_as_agent_id() -> None:
-    config = build_run_config("thread-1", None, None, assistant_id="agent-123")
-
-    assert config["configurable"]["thread_id"] == "thread-1"
-    assert config["configurable"]["agent_id"] == "agent-123"
-    assert "agent_name" not in config["configurable"]
-
-
-def test_write_agent_config_persists_skills_into_config_yaml(isolated_paths) -> None:
-    agent_id = "agent-001"
-    _write_agent_config(
-        agent_id,
-        name="中文助手",
-        description="desc",
-        skills=["s1", "s2"],
-        model="gpt-x",
-        tool_groups=["tg-a"],
-    )
-
-    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
-    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
-
-    assert raw["id"] == agent_id
-    assert raw["name"] == "中文助手"
-    assert raw["description"] == "desc"
-    assert raw["skills"] == ["s1", "s2"]
-    assert raw["model"] == "gpt-x"
-    assert raw["tool_groups"] == ["tg-a"]
-
-
-def test_load_config_extras_returns_values_written_to_yaml(isolated_paths) -> None:
-    agent_id = "agent-002"
-    _write_agent_config(
-        agent_id,
-        name="agent-002",
-        description="",
-        skills=["s1"],
-        model=None,
-        tool_groups=None,
-    )
-
-    extras = _load_config_extras(agent_id)
-    assert extras == {"skills": ["s1"], "model": None, "tool_groups": None, "common_questions": None, "template_name": None, "seat_skill_directive": None}
-
-
-def test_load_config_extras_missing_directory_returns_none_triple(isolated_paths) -> None:
-    extras = _load_config_extras("does-not-exist")
-    assert extras == {"skills": None, "model": None, "tool_groups": None, "common_questions": None, "template_name": None, "seat_skill_directive": None}
-
-
-def test_write_agent_config_persists_common_questions(isolated_paths) -> None:
-    agent_id = "agent-cq-1"
-    # Structured {title, prompt} questions round-trip as clean dict mappings.
-    questions = [
-        {"title": "能做什么", "prompt": "这个智能体能做什么？"},
-        {"title": "上传文件", "prompt": "如何上传文件？"},
-    ]
-    _write_agent_config(agent_id, name="助手", description="", common_questions=questions)
-
-    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
-    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
-    assert raw["common_questions"] == questions
-
-    extras = _load_config_extras(agent_id)
-    assert [q.model_dump() for q in extras["common_questions"]] == questions
-
-
-def test_write_agent_config_coerces_legacy_string_questions(isolated_paths) -> None:
-    """Legacy ``list[str]`` configs coerce to {title, prompt} with title=content."""
-    agent_id = "agent-cq-legacy"
-    _write_agent_config(
-        agent_id, name="助手", description="", common_questions=["如何上传文件？"]
-    )
-
-    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
-    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
-    assert raw["common_questions"] == [
-        {"title": "如何上传文件？", "prompt": "如何上传文件？"}
-    ]
-
-
-def test_write_agent_config_omits_common_questions_when_none(isolated_paths) -> None:
-    agent_id = "agent-cq-2"
-    _write_agent_config(agent_id, name="x", description="")
-
-    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
-    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
-    assert "common_questions" not in raw
-    assert _load_config_extras(agent_id)["common_questions"] is None
-
-
-def test_write_agent_config_explicit_empty_common_questions_is_persisted(isolated_paths) -> None:
-    agent_id = "agent-cq-3"
-    _write_agent_config(agent_id, name="x", description="", common_questions=[])
-
-    assert _load_config_extras(agent_id)["common_questions"] == []
-
-
-def test_write_agent_config_explicit_empty_skills_is_persisted(isolated_paths) -> None:
-    agent_id = "agent-003"
-    _write_agent_config(agent_id, name="x", description="", skills=[])
-
-    extras = _load_config_extras(agent_id)
-    assert extras["skills"] == []
-
-
-def test_write_agent_config_omits_skills_when_none(isolated_paths) -> None:
-    agent_id = "agent-004"
-    _write_agent_config(agent_id, name="x", description="")
-
-    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
-    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
-
-    assert "skills" not in raw
-    assert "model" not in raw
-    assert "tool_groups" not in raw
-
-
-@pytest.mark.asyncio
-async def test_backfill_listing_cache_mirrors_config_values() -> None:
-    store = MemoryAgentStore()
-    cfg = AgentConfig(id="a1", name="Agent", skills=["s1", "s2"], tool_groups=["g1"], model="m1")
-    await _backfill_listing_cache(store, [("a1", cfg)])
-
-    extras = await store.get_extras_for(["a1"])
-    assert extras["a1"]["skills"] == ["s1", "s2"]
-    assert extras["a1"]["tool_groups"] == ["g1"]
-    assert extras["a1"]["model"] == "m1"
-
-
-@pytest.mark.asyncio
-async def test_backfill_listing_cache_is_idempotent_and_skips_unchanged(monkeypatch) -> None:
-    store = MemoryAgentStore()
-    cfg = AgentConfig(id="a1", name="Agent", skills=["s1"], tool_groups=None, model=None)
-    await _backfill_listing_cache(store, [("a1", cfg)])
-
-    # On a second run with identical config, no re-write should occur.
-    calls = {"replace": 0, "set": 0}
-    orig_replace = store.replace_skills
-    orig_set = store.set_extras
-
-    async def _replace(agent_id, skills):
-        calls["replace"] += 1
-        await orig_replace(agent_id, skills)
-
-    async def _set(agent_id, *, tool_groups, model):
-        calls["set"] += 1
-        await orig_set(agent_id, tool_groups=tool_groups, model=model)
-
-    monkeypatch.setattr(store, "replace_skills", _replace)
-    monkeypatch.setattr(store, "set_extras", _set)
-
-    await _backfill_listing_cache(store, [("a1", cfg)])
-    assert calls == {"replace": 0, "set": 0}
-
-    # Changing the config triggers a re-write.
-    cfg2 = AgentConfig(id="a1", name="Agent", skills=["s1", "s2"], tool_groups=None, model=None)
-    await _backfill_listing_cache(store, [("a1", cfg2)])
-    assert calls["replace"] == 1
+    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
+    await store.create({"id": "private-a", "name": "Owned", "user_id": "user-a"})
+
+    assert await store.update("builtin-agent", "user-a", {"name": "Nope"}) is None
+    assert await store.update("private-a", "user-b", {"name": "Nope"}) is None
+
+    updated = await store.update("private-a", "user-a", {"name": "Renamed"})
+    assert updated is not None
+    assert updated["name"] == "Renamed"
+
+    assert await store.delete("private-a", "user-b") is False
+    assert await store.delete("private-a", "user-a") is True
+
+
+@pytest.mark.asyncio
+async def test_delete_builtin_only_targets_builtins() -> None:
+    store = MemoryAgentStore()
+
+    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
+    await store.create({"id": "owned-agent", "name": "Owned", "user_id": "user-a"})
+
+    # delete_builtin refuses to touch a user-owned agent
+    assert await store.delete_builtin("owned-agent") is False
+    assert await store.get_any("owned-agent") is not None
+
+    # delete_builtin succeeds on a built-in
+    assert await store.delete_builtin("builtin-agent") is True
+    assert await store.get_any("builtin-agent") is None
+
+    # And no-ops on a missing id
+    assert await store.delete_builtin("does-not-exist") is False
+
+
+@pytest.mark.asyncio
+async def test_update_builtin_only_targets_builtins() -> None:
+    store = MemoryAgentStore()
+
+    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
+    await store.create({"id": "owned-agent", "name": "Owned", "user_id": "user-a"})
+
+    # update_builtin succeeds on a built-in agent
+    updated = await store.update_builtin("builtin-agent", {"name": "Renamed by admin"})
+    assert updated is not None
+    assert updated["name"] == "Renamed by admin"
+
+    # update_builtin refuses to touch a user-owned agent (returns None)
+    assert await store.update_builtin("owned-agent", {"name": "Hijacked"}) is None
+
+    # And on a missing agent
+    assert await store.update_builtin("does-not-exist", {"name": "Ghost"}) is None
+
+
+@pytest.mark.asyncio
+async def test_get_editable_or_raise_admin_can_edit_builtin() -> None:
+    """Route helper: admin gets edit access to any agent; non-admin still 403."""
+    from fastapi import HTTPException
+
+    from app.gateway.routers.agents import _get_editable_or_raise
+
+    store = MemoryAgentStore()
+    await store.ensure_builtin({"id": "builtin-agent", "name": "Built in"})
+    await store.create({"id": "owned-agent", "name": "Owned", "user_id": "user-a"})
+
+    # Owner edits their own agent → returns ('owner', record)
+    record, mode = await _get_editable_or_raise(
+        store, "owned-agent", "user-a", is_admin=False
+    )
+    assert mode == "owner" and record["id"] == "owned-agent"
+
+    # Admin edits a built-in agent → returns ('admin', record)
+    record, mode = await _get_editable_or_raise(
+        store, "builtin-agent", "user-a", is_admin=True
+    )
+    assert mode == "admin" and record["id"] == "builtin-agent"
+
+    # Non-admin tries to edit a built-in → 403
+    with pytest.raises(HTTPException) as exc:
+        await _get_editable_or_raise(store, "builtin-agent", "user-a", is_admin=False)
+    assert exc.value.status_code == 403
+
+    # Admin edits someone else's custom agent → returns ('admin', record).
+    # Behavior changed: admins manage the full agent catalog (built-in + any
+    # user's agents), not just built-ins. Non-admins are still owner-only.
+    await store.create(
+        {"id": "published-other", "name": "Pub", "user_id": "user-x", "published": True}
+    )
+    record, mode = await _get_editable_or_raise(
+        store, "published-other", "user-a", is_admin=True
+    )
+    assert mode == "admin" and record["id"] == "published-other"
+
+
+def test_build_run_config_injects_custom_assistant_as_agent_id() -> None:
+    config = build_run_config("thread-1", None, None, assistant_id="agent-123")
+
+    assert config["configurable"]["thread_id"] == "thread-1"
+    assert config["configurable"]["agent_id"] == "agent-123"
+    assert "agent_name" not in config["configurable"]
+
+
+def test_write_agent_config_persists_skills_into_config_yaml(isolated_paths) -> None:
+    agent_id = "agent-001"
+    _write_agent_config(
+        agent_id,
+        name="中文助手",
+        description="desc",
+        skills=["s1", "s2"],
+        model="gpt-x",
+        tool_groups=["tg-a"],
+    )
+
+    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
+    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
+
+    assert raw["id"] == agent_id
+    assert raw["name"] == "中文助手"
+    assert raw["description"] == "desc"
+    assert raw["skills"] == ["s1", "s2"]
+    assert raw["model"] == "gpt-x"
+    assert raw["tool_groups"] == ["tg-a"]
+
+
+def test_load_config_extras_returns_values_written_to_yaml(isolated_paths) -> None:
+    agent_id = "agent-002"
+    _write_agent_config(
+        agent_id,
+        name="agent-002",
+        description="",
+        skills=["s1"],
+        model=None,
+        tool_groups=None,
+    )
+
+    extras = _load_config_extras(agent_id)
+    assert extras == {
+        "skills": ["s1"],
+        "model": None,
+        "tool_groups": None,
+        "common_questions": None,
+        "template_name": None,
+        "seat_skill_directive": None,
+        "job_responsibilities": None,
+        "llmwiki_knowledge_base_ids": None,
+    }
+
+
+def test_load_config_extras_missing_directory_returns_none_triple(isolated_paths) -> None:
+    extras = _load_config_extras("does-not-exist")
+    assert extras == {
+        "skills": None,
+        "model": None,
+        "tool_groups": None,
+        "common_questions": None,
+        "template_name": None,
+        "seat_skill_directive": None,
+        "job_responsibilities": None,
+        "llmwiki_knowledge_base_ids": None,
+    }
+
+
+def test_write_agent_config_persists_common_questions(isolated_paths) -> None:
+    agent_id = "agent-cq-1"
+    # Structured {title, prompt} questions round-trip as clean dict mappings.
+    questions = [
+        {"title": "能做什么", "prompt": "这个智能体能做什么？"},
+        {"title": "上传文件", "prompt": "如何上传文件？"},
+    ]
+    _write_agent_config(agent_id, name="助手", description="", common_questions=questions)
+
+    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
+    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
+    assert raw["common_questions"] == questions
+
+    extras = _load_config_extras(agent_id)
+    assert [q.model_dump() for q in extras["common_questions"]] == questions
+
+
+def test_write_agent_config_coerces_legacy_string_questions(isolated_paths) -> None:
+    """Legacy ``list[str]`` configs coerce to {title, prompt} with title=content."""
+    agent_id = "agent-cq-legacy"
+    _write_agent_config(
+        agent_id, name="助手", description="", common_questions=["如何上传文件？"]
+    )
+
+    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
+    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
+    assert raw["common_questions"] == [
+        {"title": "如何上传文件？", "prompt": "如何上传文件？"}
+    ]
+
+
+def test_write_agent_config_omits_common_questions_when_none(isolated_paths) -> None:
+    agent_id = "agent-cq-2"
+    _write_agent_config(agent_id, name="x", description="")
+
+    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
+    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
+    assert "common_questions" not in raw
+    assert _load_config_extras(agent_id)["common_questions"] is None
+
+
+def test_write_agent_config_explicit_empty_common_questions_is_persisted(isolated_paths) -> None:
+    agent_id = "agent-cq-3"
+    _write_agent_config(agent_id, name="x", description="", common_questions=[])
+
+    assert _load_config_extras(agent_id)["common_questions"] == []
+
+
+def test_write_agent_config_explicit_empty_skills_is_persisted(isolated_paths) -> None:
+    agent_id = "agent-003"
+    _write_agent_config(agent_id, name="x", description="", skills=[])
+
+    extras = _load_config_extras(agent_id)
+    assert extras["skills"] == []
+
+
+def test_write_agent_config_omits_skills_when_none(isolated_paths) -> None:
+    agent_id = "agent-004"
+    _write_agent_config(agent_id, name="x", description="")
+
+    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
+    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
+
+    assert "skills" not in raw
+    assert "model" not in raw
+    assert "tool_groups" not in raw
+
+
+def test_write_agent_config_persists_job_responsibilities(isolated_paths) -> None:
+    agent_id = "agent-job-1"
+    _write_agent_config(
+        agent_id,
+        name="助手",
+        description="",
+        job_responsibilities="  负责意图识别与业务链确认  ",
+    )
+
+    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
+    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
+    assert raw["job_responsibilities"] == "负责意图识别与业务链确认"
+    assert _load_config_extras(agent_id)["job_responsibilities"] == "负责意图识别与业务链确认"
+
+
+def test_write_agent_config_omits_blank_job_responsibilities(isolated_paths) -> None:
+    agent_id = "agent-job-2"
+    _write_agent_config(agent_id, name="助手", description="", job_responsibilities="   ")
+
+    config_file = paths_module.get_paths().agent_dir(agent_id) / "config.yaml"
+    raw = yaml.safe_load(config_file.read_text(encoding="utf-8"))
+    assert "job_responsibilities" not in raw
+    assert _load_config_extras(agent_id)["job_responsibilities"] is None
+
+
+@pytest.mark.asyncio
+async def test_backfill_listing_cache_mirrors_config_values() -> None:
+    store = MemoryAgentStore()
+    cfg = AgentConfig(id="a1", name="Agent", skills=["s1", "s2"], tool_groups=["g1"], model="m1")
+    await _backfill_listing_cache(store, [("a1", cfg)])
+
+    extras = await store.get_extras_for(["a1"])
+    assert extras["a1"]["skills"] == ["s1", "s2"]
+    assert extras["a1"]["tool_groups"] == ["g1"]
+    assert extras["a1"]["model"] == "m1"
+
+
+@pytest.mark.asyncio
+async def test_backfill_listing_cache_is_idempotent_and_skips_unchanged(monkeypatch) -> None:
+    store = MemoryAgentStore()
+    cfg = AgentConfig(id="a1", name="Agent", skills=["s1"], tool_groups=None, model=None)
+    await _backfill_listing_cache(store, [("a1", cfg)])
+
+    # On a second run with identical config, no re-write should occur.
+    calls = {"replace": 0, "set": 0}
+    orig_replace = store.replace_skills
+    orig_set = store.set_extras
+
+    async def _replace(agent_id, skills):
+        calls["replace"] += 1
+        await orig_replace(agent_id, skills)
+
+    async def _set(agent_id, *, tool_groups, model):
+        calls["set"] += 1
+        await orig_set(agent_id, tool_groups=tool_groups, model=model)
+
+    monkeypatch.setattr(store, "replace_skills", _replace)
+    monkeypatch.setattr(store, "set_extras", _set)
+
+    await _backfill_listing_cache(store, [("a1", cfg)])
+    assert calls == {"replace": 0, "set": 0}
+
+    # Changing the config triggers a re-write.
+    cfg2 = AgentConfig(id="a1", name="Agent", skills=["s1", "s2"], tool_groups=None, model=None)
+    await _backfill_listing_cache(store, [("a1", cfg2)])
+    assert calls["replace"] == 1