接口联调

frontend-web/src/roundtable-planning/api/intent.ts

@@ -0,0 +1,186 @@
+import { apiFetch } from "@/core/api/fetch-client";
+import { getBackendBaseURL } from "@/core/config";
+
+/**
+ * Task-intent agent (Step 1) API. Mirror of multi-agent.ts but with a single
+ * built-in agent and no per-call coordinator: one /init creates a thread
+ * tied to the built-in `roundtable-intent` agent, and each /stream POST is
+ * one user-turn of the clarification dialogue. The terminal status frame
+ * carries either `asking` (more turns needed) or `done` (structured summary
+ * available — hand off to Step 2).
+ */
+
+export interface IntentInitResponse {
+  status: "success";
+  agent_id: string;
+  thread_id: string;
+}
+
+/**
+ * Create a fresh thread for an intent-clarification session.
+ * The backend returns the built-in agent id so the frontend doesn't hardcode it.
+ */
+export async function initIntent(model?: string): Promise<IntentInitResponse> {
+  const res = await apiFetch(`${getBackendBaseURL()}/api/intent/init`, {
+    method: "POST",
+    headers: { "Content-Type": "application/json" },
+    body: JSON.stringify({ model }),
+  });
+  if (!res.ok) {
+    const text = await res.text().catch(() => "");
+    throw new Error(`init intent failed (${res.status}): ${text}`);
+  }
+  return (await res.json()) as IntentInitResponse;
+}
+
+/**
+ * Structured task summary the intent agent emits when it judges the user's
+ * task is fully specified. The frontend feeds this into Step 2's initial
+ * leader prompt.
+ */
+export interface IntentSummary {
+  /** One-line statement of what the user wants accomplished. */
+  objective: string;
+  /** Hard constraints (budget / time / compliance / scope). At least 1 entry. */
+  constraints: string[];
+  /** Optional assumptions the agent is making on the user's behalf. */
+  assumptions?: string[];
+}
+
+export interface StreamIntentRequest {
+  threadId: string;
+  message: string;
+  model?: string;
+  signal?: AbortSignal;
+  /** Fires for every AI text chunk parsed out of LangGraph messages-tuple frames. */
+  onTextDelta?: (text: string, messageId: string) => void;
+}
+
+export interface StreamIntentResult {
+  /** "asking" → keep the dialogue going; "done" → summary is ready; "error" → backend failure. */
+  status: "asking" | "done" | "error";
+  /** Final assistant text for this turn (markdown, may contain the [INTENT_READY] block). */
+  content: string;
+  /** Populated only when status === "done". */
+  summary: IntentSummary | null;
+}
+
+/**
+ * Recognise the terminal status frame the backend appends after the upstream
+ * LangGraph stream finishes. Must carry `status` and `content`; `summary`
+ * is optional.
+ */
+function isFinalIntentFrame(value: unknown): value is {
+  status: "asking" | "done" | "error";
+  content?: string;
+  summary?: IntentSummary | null;
+} {
+  if (!value || typeof value !== "object") return false;
+  if (!("status" in value) || !("content" in value)) return false;
+  const status = (value as { status: unknown }).status;
+  return status === "asking" || status === "done" || status === "error";
+}
+
+/**
+ * Reuse the same delta-extraction shape as multi-agent.ts so AI text streams
+ * smoothly into the frontend bubble. Returns null for tool / value / tool
+ * result frames.
+ */
+function extractAiTextDelta(parsed: unknown): { text: string; id: string } | null {
+  if (!Array.isArray(parsed) || parsed.length < 1) return null;
+  const chunk = parsed[0];
+  if (!chunk || typeof chunk !== "object") return null;
+  const obj = chunk as Record<string, unknown>;
+  const type = String(obj.type ?? "").toLowerCase();
+  if (type !== "ai" && type !== "aimessagechunk" && type !== "aimessage") return null;
+  const content = obj.content;
+  let text = "";
+  if (typeof content === "string") {
+    text = content;
+  } else if (Array.isArray(content)) {
+    for (const item of content) {
+      if (typeof item === "string") text += item;
+      else if (item && typeof item === "object") {
+        const block = item as Record<string, unknown>;
+        const blockText = block.text;
+        if (typeof blockText === "string") text += blockText;
+      }
+    }
+  }
+  if (!text) return null;
+  const id = typeof obj.id === "string" ? obj.id : "";
+  return { text, id };
+}
+
+/**
+ * Open an SSE connection to /api/intent/stream and resolve when the upstream
+ * stream completes. Each call is one user turn; reuse the same threadId
+ * across calls to keep the dialogue history.
+ */
+export async function streamIntent(req: StreamIntentRequest): Promise<StreamIntentResult> {
+  const res = await apiFetch(`${getBackendBaseURL()}/api/intent/stream`, {
+    method: "POST",
+    headers: { "Content-Type": "application/json" },
+    body: JSON.stringify({
+      thread_id: req.threadId,
+      message: req.message,
+      model: req.model,
+    }),
+    signal: req.signal,
+  });
+
+  if (!res.ok || !res.body) {
+    const text = await res.text().catch(() => "");
+    throw new Error(`intent/stream failed (${res.status}): ${text}`);
+  }
+
+  const reader = res.body.getReader();
+  const decoder = new TextDecoder();
+  let buffer = "";
+  let lastEvent = "";
+
+  const result: StreamIntentResult = { status: "asking", content: "", summary: null };
+
+  while (true) {
+    const { done, value } = await reader.read();
+    if (done) break;
+    buffer += decoder.decode(value, { stream: true });
+    const lines = buffer.split("\n");
+    buffer = lines.pop() ?? "";
+
+    for (const rawLine of lines) {
+      const line = rawLine.trimEnd();
+      if (!line) continue;
+      if (line.startsWith("event:")) {
+        lastEvent = line.slice(6).trim();
+        continue;
+      }
+      if (!line.startsWith("data:")) continue;
+      const payload = line.slice(5).trim();
+      if (!payload) continue;
+
+      let parsed: unknown;
+      try {
+        parsed = JSON.parse(payload);
+      } catch {
+        continue;
+      }
+
+      if (isFinalIntentFrame(parsed)) {
+        result.status = parsed.status;
+        if (typeof parsed.content === "string") result.content = parsed.content;
+        if (parsed.summary && typeof parsed.summary === "object") {
+          result.summary = parsed.summary as IntentSummary;
+        }
+        continue;
+      }
+
+      if (lastEvent === "messages") {
+        const delta = extractAiTextDelta(parsed);
+        if (delta) req.onTextDelta?.(delta.text, delta.id);
+      }
+    }
+  }
+
+  return result;
+}

frontend-web/src/roundtable-planning/pages/RoundtablePlanningPage.tsx

@@ -13,6 +13,7 @@ import {
   type StreamPhase,
   type ThreadIdMap,
 } from "@/roundtable-planning/api/multi-agent";
+import { initIntent, streamIntent, type IntentSummary } from "@/roundtable-planning/api/intent";
 import { 
   Shield, 
   MessageSquare, 
@@ -305,6 +306,17 @@ export default function RoundtablePlanningPage() {
   // Collapsing gives the conversation log more vertical space.
   const [isUpperPanelCollapsed, setIsUpperPanelCollapsed] = useState<boolean>(false);
 
+  // Step 1 intent-clarification state (lives only for one step-1 visit).
+  // The thread is created lazily on first entry; intentReady is the
+  // structured summary returned by the intent agent when it judges the
+  // task is fully specified — Step 2's initial leader prompt is built
+  // from this.
+  const [intentThreadId, setIntentThreadId] = useState<string | null>(null);
+  const [intentReady, setIntentReady] = useState<IntentSummary | null>(null);
+  const [intentInitError, setIntentInitError] = useState<string | null>(null);
+  const [isIntentSending, setIsIntentSending] = useState<boolean>(false);
+  const step1InitRef = useRef<boolean>(false);
+
   // Multi-agent orchestration state (lives only for one step-2 visit).
   const [threadIds, setThreadIds] = useState<ThreadIdMap | null>(null);
   const [coordinatorName, setCoordinatorName] = useState<string | null>(null);
@@ -514,14 +526,32 @@ export default function RoundtablePlanningPage() {
     );
   }, []);
 
-  // Pull the consolidated user intent out of step-1 history. Falls back to a
-  // generic prompt if the user never spoke.
+  // Build the Step 2 leader's initial prompt. Prefer the intent agent's
+  // structured summary when available — it already encodes objective +
+  // constraints + assumptions. Fall back to "last user message + budget"
+  // for sessions that bypassed Step 1 or hit an init error.
   const buildInitialIntent = useCallback(() => {
+    if (intentReady) {
+      const parts: string[] = [intentReady.objective];
+      if (intentReady.constraints?.length) {
+        parts.push("");
+        parts.push("约束条件:");
+        for (const c of intentReady.constraints) parts.push(`- ${c}`);
+      }
+      if (intentReady.assumptions?.length) {
+        parts.push("");
+        parts.push("关键假设:");
+        for (const a of intentReady.assumptions) parts.push(`- ${a}`);
+      }
+      parts.push("");
+      parts.push("用户已经过任务理解智能体澄清,以上信息可以直接作为研讨依据。请基于上述目标与约束,在圆桌内调度合适的子智能体完成研讨,不要再反问用户。");
+      return parts.join("\n");
+    }
     const lastUserMessage = [...interactiveMessages].reverse().find(m => m.sender === "user");
     const userIntent = lastUserMessage?.content || "用户尚未给出明确意图";
     const constraints = `预算硬约束:${budgetLimit} 万元`;
     return `${userIntent}\n\n约束条件:${constraints}\n\n请基于上述目标与约束,在圆桌内调度合适的子智能体完成研讨。`;
-  }, [interactiveMessages, budgetLimit]);
+  }, [interactiveMessages, budgetLimit, intentReady]);
 
   // Run one full orchestration step: 1) leader call → 2) dispatch specials in
   // parallel → 3) recurse with a synthesis prompt until leader stops dispatching.
@@ -868,49 +898,111 @@ export default function RoundtablePlanningPage() {
     setShowToast(msg);
   };
 
-  // Chat sender logic for Step 1 interaction
-  const handleSendMessage = () => {
-    if (!inputText.trim()) return;
+  // Chat sender logic for Step 1 interaction — drives the real intent agent
+  // via /api/intent/stream. The agent streams its reply into a placeholder
+  // bubble; if its final status is "done" we capture the structured summary
+  // for the Step 2 handoff.
+  const handleSendMessage = async () => {
+    const userText = inputText.trim();
+    if (!userText) return;
+    if (!intentThreadId) {
+      triggerToast("⚠️ 任务理解智能体尚未就绪,请稍候");
+      return;
+    }
+    if (isIntentSending) {
+      triggerToast("⏳ 智能体正在回复,请稍候");
+      return;
+    }
 
-    const userMsgTime = new Date().toLocaleTimeString("zh-CN", { hour12: false });
+    const userMsgTime = nowTime();
+    const userMsgId = Date.now() + Math.random();
     const userMsg = {
-      id: interactiveMessages.length + 1,
+      id: userMsgId,
       sender: "user",
       senderName: "用户",
-      content: inputText,
+      content: userText,
       time: userMsgTime,
     };
-
-    // Append user message immediately
-    const updatedMessages = [...interactiveMessages, userMsg];
-    setInteractiveMessages(updatedMessages);
+    setInteractiveMessages(prev => [...prev, userMsg]);
     setInputText("");
     setCharactersLeft(0);
 
-    // Update Clarification timeline list on the right beautifully!
-    const newTimelineEntry = {
-      id: timelineEntries.length + 1,
-      time: userMsgTime,
-      title: "用户实时反馈",
-      desc: userMsg.content.length > 25 ? userMsg.content.substring(0, 25) + "..." : userMsg.content
-    };
-    setTimelineEntries(prev => [...prev, newTimelineEntry]);
+    // Right-side timeline gets a short summary too.
+    setTimelineEntries(prev => [
+      ...prev,
+      {
+        id: prev.length + 1,
+        time: userMsgTime,
+        title: "用户实时反馈",
+        desc: userText.length > 25 ? userText.substring(0, 25) + "..." : userText,
+      },
+    ]);
 
-    triggerToast("消息发送成功！任务意图分析中...");
-
-    // Simulated responsive intelligence robot typing back
-    setTimeout(() => {
-      const responseTime = new Date().toLocaleTimeString("zh-CN", { hour12: false });
-      const automatedReplyLine = {
-        id: updatedMessages.length + 2,
+    // Placeholder agent bubble — text deltas accumulate into it.
+    const agentMsgId = Date.now() + Math.random() + 0.01;
+    setInteractiveMessages(prev => [
+      ...prev,
+      {
+        id: agentMsgId,
         sender: "agent",
         senderName: "任务理解智能体",
-        content: `已收到关于“${userMsg.content}”的更新！我已自动校对预算配比与合规规范。当前规划框架可信度提升 4.2%。请您继续审视下方待确认事项，或随时确认以进入多智能体圆桌研讨。`,
-        time: responseTime,
-      };
-      setInteractiveMessages(prev => [...prev, automatedReplyLine]);
-      triggerToast("任务理解智能体已回复！");
-    }, 1500);
+        content: "",
+        time: nowTime(),
+        streaming: true,
+      } as any,
+    ]);
+
+    setIsIntentSending(true);
+    try {
+      const res = await streamIntent({
+        threadId: intentThreadId,
+        message: userText,
+        model: DEFAULT_MODEL,
+        onTextDelta: (delta) => {
+          setInteractiveMessages(prev =>
+            prev.map(m => (m.id === agentMsgId ? { ...m, content: (m.content ?? "") + delta } : m)),
+          );
+        },
+      });
+      // Finalize the bubble with the authoritative final text from the
+      // status frame (covers cases where deltas didn't accumulate cleanly).
+      setInteractiveMessages(prev =>
+        prev.map(m =>
+          m.id === agentMsgId
+            ? { ...m, content: res.content || m.content || "(无文本输出)", streaming: false } as any
+            : m,
+        ),
+      );
+      if (res.status === "done" && res.summary) {
+        setIntentReady(res.summary);
+        setTimelineEntries(prev => [
+          ...prev,
+          {
+            id: prev.length + 1,
+            time: nowTime(),
+            title: "任务意图已明确",
+            desc: res.summary!.objective.length > 25
+              ? res.summary!.objective.substring(0, 25) + "..."
+              : res.summary!.objective,
+          },
+        ]);
+        triggerToast("✅ 任务理解完成,可进入多智能体圆桌研讨");
+      } else if (res.status === "error") {
+        triggerToast(`❌ 智能体调用失败: ${res.content}`);
+      }
+    } catch (err) {
+      console.error("intent stream failed:", err);
+      setInteractiveMessages(prev =>
+        prev.map(m =>
+          m.id === agentMsgId
+            ? { ...m, content: `(错误) ${(err as Error).message}`, streaming: false } as any
+            : m,
+        ),
+      );
+      triggerToast(`❌ 发送失败: ${(err as Error).message}`);
+    } finally {
+      setIsIntentSending(false);
+    }
   };
 
   // Skip step logic
@@ -939,20 +1031,88 @@ export default function RoundtablePlanningPage() {
     triggerToast("追问生成就绪！");
   };
 
-  // Step 2 role statuses transition when entering roundtable!
-  // On Step 2 entry we set every agent seat to "正在上线" (spinner state);
-  // the /init SSE stream then flips each one to "在线" as its thread/agent
-  // actually finishes on the backend. Seat 1 (human commander) stays online
-  // throughout — it's not part of the bring-up.
+  // Step 1 init effect: create a fresh intent-agent thread the first time
+  // the user lands on step 1 (and on every re-entry, since clearing thread
+  // history is desirable here — each new planning session starts clean).
+  //
+  // Note on React 18 StrictMode: effects run twice in dev. We reset
+  // `step1InitRef.current` to false in the cleanup so the second mount can
+  // actually run init — otherwise the ref stays true, the second effect
+  // returns early, and the first effect's `cancelled` flag prevents its
+  // result from setting state. Net effect would be a successful API call
+  // whose result never lands and the Send button stays disabled.
+  useEffect(() => {
+    if (currentStep !== 1) {
+      step1InitRef.current = false;
+      return;
+    }
+    if (step1InitRef.current) return;
+    step1InitRef.current = true;
+
+    let cancelled = false;
+    (async () => {
+      setIntentReady(null);
+      setIntentInitError(null);
+      setIntentThreadId(null);
+      setTimelineEntries([]);
+      setInteractiveMessages([
+        {
+          id: 1,
+          sender: "agent",
+          senderName: "任务理解智能体",
+          content: "你好,我是任务理解智能体。请用 1-2 句话告诉我你这次想做什么,以及有哪些预算/时间/合规上的硬约束?",
+          time: nowTime(),
+        },
+      ]);
+      try {
+        const res = await initIntent(DEFAULT_MODEL);
+        if (cancelled) return;
+        setIntentThreadId(res.thread_id);
+        setTimelineEntries([
+          {
+            id: 1,
+            time: nowTime(),
+            title: "任务理解智能体已就绪",
+            desc: `agent_id=${res.agent_id}`,
+          },
+        ]);
+      } catch (err) {
+        if (cancelled) return;
+        console.error("init intent failed:", err);
+        setIntentInitError((err as Error).message);
+      }
+    })();
+    return () => {
+      cancelled = true;
+      step1InitRef.current = false;
+    };
+    // eslint-disable-next-line react-hooks/exhaustive-deps
+  }, [currentStep]);
+
+  // Role panel transitions:
+  // - Step 1 shows ONLY the commander + the task-intent agent (2 entries).
+  //   The intent agent flips to 在线 once /api/intent/init returns; if it
+  //   errors we mark 异常 so the user knows why send is disabled.
+  // - Step 2 lights up all 7 seats (each flipped 在线 as init SSE reports).
+  // - Step 3 marks the whole roster as 已完成.
   useEffect(() => {
     if (currentStep === 1) {
-      setRoles(makeRoles("等待激活", { 2: "在线" }));
+      setRoles([
+        { id: 1, name: "指挥员 / 主持人", type: "human", status: "在线" },
+        {
+          id: 100,
+          name: "任务理解",
+          agentId: "roundtable-intent",
+          type: "agent",
+          status: intentInitError ? "异常" : intentThreadId ? "在线" : "正在上线",
+        },
+      ]);
     } else if (currentStep === 2) {
       setRoles(makeRoles("正在上线", { 1: "在线" }));
     } else {
       setRoles(makeRoles("已完成"));
     }
-  }, [currentStep]);
+  }, [currentStep, intentThreadId, intentInitError]);
 
   // Inject user input to the roundtable room — sends it as a new leader call.
   const handleInjectToRoundtable = () => {
@@ -1426,157 +1586,95 @@ export default function RoundtablePlanningPage() {
                     任务意图理解 (人机问答)
                   </h3>
                 </div>
+                {/* Real turn counter — counts user messages in the live
+                    interactiveMessages list. "X 轮" instead of "X/3" because
+                    we don't bound the number of clarification turns. */}
                 <span className="text-[11px] font-bold text-slate-400 bg-white border px-2 py-0.5 rounded-md">
-                  问答轮次：3/3
+                  问答轮次：{interactiveMessages.filter(m => m.sender === "user").length} 轮
                 </span>
               </div>
 
               {/* Chat list viewport */}
               <div className="flex-1 overflow-y-auto p-6 space-y-4 font-normal">
                 
-                {/* Message 1 (User / 用户) */}
-                <div id="msg-item-1" className="flex items-start justify-between gap-3 border-b border-slate-50 pb-3">
-                  <div className="flex items-start gap-3">
-                    <Man1Avatar />
-                    <div className="space-y-1">
-                      <div className="flex items-baseline gap-2">
-                        <span className="text-xs font-bold text-primary">用户</span>
+                {/* Step 1 dialogue is rendered by `interactiveMessages.map()`
+                    inside the 问答交互区 sub-block below. The static demo
+                    bubbles that previously lived here have been removed. */}
+
+                {/* --- INTENT SUMMARY BOX --- emerald (ready) / slate (waiting) ---
+                    Rendered from `intentReady` when the agent has emitted a
+                    structured summary; otherwise shows a placeholder with
+                    a disabled "进入下一步" button so the user knows what they
+                    need to wait for. */}
+                {intentReady ? (
+                  <div className="border border-emerald-200 bg-[#f0fdf4]/60 rounded-2xl p-4.5 shadow-[0_2px_12px_rgba(16,185,129,0.04)]">
+                    <div className="flex items-center justify-between gap-3 border-b border-emerald-100 pb-3 mb-3.5">
+                      <div className="flex items-center gap-2">
+                        <CheckCircle className="h-5 w-5 text-emerald-500 fill-emerald-50" strokeWidth={2.5} />
+                        <span className="text-sm font-extrabold text-slate-900">
+                          已形成明确任务意图
+                        </span>
                       </div>
-                      <p className="text-xs text-slate-700 leading-relaxed max-w-[500px]">
-                        请帮我围绕某复杂任务制定详细规划方案。
-                      </p>
+                      <button
+                        id="confirm-roundtable-btn"
+                        onClick={() => {
+                          setCurrentStep(2);
+                          triggerToast("成功确认！已切换至第二阶段圆桌研讨系统");
+                        }}
+                        className="bg-primary hover:bg-[var(--brand-900)] text-white font-bold text-xs px-4 py-2 rounded-xl transition shadow-[0_2px_8px_rgba(245,34,45,0.2)] active:scale-95 shrink-0"
+                      >
+                        用户确认，进入多智能体研讨
+                      </button>
                     </div>
-                  </div>
-                  <span className="text-[10px] text-slate-400 font-mono shrink-0">10:32:15</span>
-                </div>
 
-                {/* Message 2 (Agent) */}
-                <div id="msg-item-2" className="flex items-start justify-between gap-3 border-b border-slate-50 pb-3">
-                  <div className="flex items-start gap-3">
-                    <RobotAvatar bg="bg-[var(--brand-200)]/50" />
-                    <div className="space-y-1">
-                      <div className="flex items-baseline gap-2">
-                        <span className="text-xs font-bold text-[var(--brand-900)]">任务理解智能体</span>
+                    <div className="grid grid-cols-1 md:grid-cols-3 gap-3">
+                      <div className="space-y-0.5 md:col-span-3">
+                        <span className="text-[10px] font-bold text-slate-400 block tracking-tight">任务目标</span>
+                        <p className="text-xs font-bold text-slate-800 leading-snug">{intentReady.objective}</p>
                       </div>
-                      <p className="text-xs text-slate-700 leading-relaxed max-w-[500px] font-medium bg-primary/5 px-3 py-2 rounded-xl border border-primary/10">
-                        请明确目标、时间要求与资源限制。
-                      </p>
-                    </div>
-                  </div>
-                  <span className="text-[10px] text-slate-400 font-mono shrink-0">10:32:22</span>
-                </div>
-
-                {/* Message 3 (User / 用户) */}
-                <div id="msg-item-3" className="flex items-start justify-between gap-3 border-b border-slate-50 pb-3">
-                  <div className="flex items-start gap-3">
-                    <Man1Avatar />
-                    <div className="space-y-1">
-                      <div className="flex items-baseline gap-2">
-                        <span className="text-xs font-bold text-primary">用户</span>
+                      <div className="space-y-1 md:col-span-2">
+                        <span className="text-[10px] font-bold text-slate-400 block tracking-tight">硬约束</span>
+                        <ul className="text-xs font-bold text-slate-800 leading-snug list-disc pl-4 space-y-0.5">
+                          {intentReady.constraints.map((c, i) => (
+                            <li key={i}>{c}</li>
+                          ))}
+                        </ul>
                       </div>
-                      <p className="text-xs text-slate-700 leading-relaxed max-w-[500px]">
-                        72小时内完成，可用预算500万，人员30人以内。
-                      </p>
+                      {intentReady.assumptions && intentReady.assumptions.length > 0 && (
+                        <div className="space-y-1">
+                          <span className="text-[10px] font-bold text-slate-400 block tracking-tight">关键假设</span>
+                          <ul className="text-xs font-medium text-slate-600 leading-snug list-disc pl-4 space-y-0.5">
+                            {intentReady.assumptions.map((a, i) => (
+                              <li key={i}>{a}</li>
+                            ))}
+                          </ul>
+                        </div>
+                      )}
                     </div>
                   </div>
-                  <span className="text-[10px] text-slate-400 font-mono shrink-0">10:33:02</span>
-                </div>
-
-                {/* Message 4 (Agent) */}
-                <div id="msg-item-4" className="flex items-start justify-between gap-3 border-b border-slate-50 pb-3">
-                  <div className="flex items-start gap-3">
-                    <RobotAvatar bg="bg-[var(--brand-200)]/50" />
-                    <div className="space-y-1">
-                      <div className="flex items-baseline gap-2">
-                        <span className="text-xs font-bold text-[var(--brand-900)]">任务理解智能体</span>
+                ) : (
+                  <div className="border border-slate-200 bg-slate-50/60 rounded-2xl p-4.5">
+                    <div className="flex items-center justify-between gap-3">
+                      <div className="flex items-center gap-2">
+                        <Loader2 className="h-5 w-5 text-slate-400 animate-spin" />
+                        <span className="text-sm font-bold text-slate-700">
+                          等待任务理解智能体澄清完成...
+                        </span>
                       </div>
-                      <p className="text-xs text-slate-700 leading-relaxed max-w-[500px] font-medium bg-primary/5 px-3 py-2 rounded-xl border border-primary/10">
-                        是否优先保证主方案可执行性，并同时给出保守备选方案？
+                      <button
+                        disabled
+                        className="bg-slate-200 text-slate-400 font-bold text-xs px-4 py-2 rounded-xl shrink-0 cursor-not-allowed"
+                      >
+                        用户确认，进入多智能体研讨
+                      </button>
+                    </div>
+                    {intentInitError && (
+                      <p className="mt-2 text-[11px] text-red-500 font-medium">
+                        智能体初始化失败: {intentInitError}
                       </p>
-                    </div>
+                    )}
                   </div>
-                  <span className="text-[10px] text-slate-400 font-mono shrink-0">10:33:15</span>
-                </div>
-
-                {/* Message 5 (User / 用户) */}
-                <div id="msg-item-5" className="flex items-start justify-between gap-3 border-b border-slate-50 pb-3">
-                  <div className="flex items-start gap-3">
-                    <Man1Avatar />
-                    <div className="space-y-1">
-                      <div className="flex items-baseline gap-2">
-                        <span className="text-xs font-bold text-primary">用户</span>
-                      </div>
-                      <p className="text-xs text-slate-700 leading-relaxed max-w-[500px]">
-                        是，请保留主方案并给出备选方案。
-                      </p>
-                    </div>
-                  </div>
-                  <span className="text-[10px] text-slate-400 font-mono shrink-0">10:33:36</span>
-                </div>
-
-                {/* Message 6 (Agent) */}
-                <div id="msg-item-6" className="flex items-start justify-between gap-3 pb-2">
-                  <div className="flex items-start gap-3">
-                    <RobotAvatar bg="bg-[var(--brand-200)]/50" />
-                    <div className="space-y-1">
-                      <div className="flex items-baseline gap-2">
-                        <span className="text-xs font-bold text-[var(--brand-900)]">任务理解智能体</span>
-                      </div>
-                      <p className="text-xs text-slate-700 leading-relaxed max-w-[500px] font-medium bg-primary/5 px-3 py-2 rounded-xl border border-primary/10">
-                        已基本形成任务意图，请确认。
-                      </p>
-                    </div>
-                  </div>
-                  <span className="text-[10px] text-slate-400 font-mono shrink-0">10:33:52</span>
-                </div>
-
-                {/* --- GREEN INTENT BOX (1-to-1) --- */}
-                <div className="border border-emerald-200 bg-[#f0fdf4]/60 rounded-2xl p-4.5 shadow-[0_2px_12px_rgba(16,185,129,0.04)]">
-                  {/* Top row */}
-                  <div className="flex items-center justify-between gap-3 border-b border-emerald-100 pb-3 mb-3.5">
-                    <div className="flex items-center gap-2">
-                      <CheckCircle className="h-5 w-5 text-emerald-500 fill-emerald-50" strokeWidth={2.5} />
-                      <span className="text-sm font-extrabold text-slate-900">
-                        已形成明确任务意图
-                      </span>
-                    </div>
-
-                    <button
-                      id="confirm-roundtable-btn"
-                      onClick={() => {
-                        setCurrentStep(2);
-                        triggerToast("成功确认！已切换至第二阶段圆桌研讨系统");
-                      }}
-                      className="bg-primary hover:bg-[var(--brand-900)] text-white font-bold text-xs px-4 py-2 rounded-xl transition shadow-[0_2px_8px_rgba(245,34,45,0.2)] active:scale-95 shrink-0"
-                    >
-                      用户确认，进入多智能体研讨
-                    </button>
-                  </div>
-
-                  {/* Structured properties in columns */}
-                  <div className="grid grid-cols-2 md:grid-cols-5 gap-3">
-                    <div className="space-y-0.5">
-                      <span className="text-[10px] font-bold text-slate-400 block tracking-tight">目标</span>
-                      <p className="text-xs font-bold text-slate-800 leading-tight">制定可执行的<br className="hidden md:inline" />详细规划方案</p>
-                    </div>
-                    <div className="space-y-0.5">
-                      <span className="text-[10px] font-bold text-slate-400 block tracking-tight">时间边界</span>
-                      <p className="text-xs font-bold text-slate-800 leading-tight">72小时内完成</p>
-                    </div>
-                    <div className="space-y-0.5">
-                      <span className="text-[10px] font-bold text-slate-400 block tracking-tight">资源约束</span>
-                      <p className="text-xs font-bold text-slate-800 leading-tight">预算 ≤ 500万，<br />人员 ≤ 30人</p>
-                    </div>
-                    <div className="space-y-0.5">
-                      <span className="text-[10px] font-bold text-slate-400 block tracking-tight">优先级</span>
-                      <p className="text-xs font-bold text-slate-800 leading-tight">主方案可执行性优先，<br />兼顾备选方案</p>
-                    </div>
-                    <div className="space-y-0.5 col-span-2 md:col-span-1">
-                      <span className="text-[10px] font-bold text-slate-400 block tracking-tight">待确认事项</span>
-                      <p className="text-xs font-bold text-slate-800 leading-normal">备选方案范围、<br />预算触发条件</p>
-                    </div>
-                  </div>
-                </div>
+                )}
 
                 {/* --- Q&A INTERACTION AREA (问答交互区) --- */}
                 <div className="border border-red-100 rounded-2xl bg-slate-50/20 shadow-[0_2px_12px_rgba(245,34,45,0.01)] overflow-hidden">
@@ -1594,26 +1692,41 @@ export default function RoundtablePlanningPage() {
 
                   {/* Content dialogue lines inside interaction */}
                   <div className="p-4.5 space-y-3.5 max-h-[280px] overflow-y-auto">
-                    {interactiveMessages.map((msg) => (
-                      <div key={msg.id} className="flex items-start justify-between gap-3 border-b border-slate-100/50 pb-2.5 last:border-0 last:pb-0">
-                        <div className="flex items-start gap-2.5">
-                          {msg.sender === "user" ? (
-                            <Man1Avatar size="h-6 w-6" />
-                          ) : (
-                            <RobotAvatar bg="bg-[var(--brand-200)]/50" size="h-6 w-6" />
-                          )}
-                          <div>
-                            <span className={`text-[10.5px] font-bold block mb-0.5 ${msg.sender === 'user' ? 'text-primary' : 'text-slate-700'}`}>
-                              {msg.senderName}
-                            </span>
-                            <span className="text-xs text-slate-700 leading-normal font-normal">
-                              {msg.content}
-                            </span>
+                    {interactiveMessages.map((msg) => {
+                      const isStreaming = (msg as { streaming?: boolean }).streaming === true;
+                      const hasContent = (msg.content ?? "").length > 0;
+                      return (
+                        <div key={msg.id} className="flex items-start justify-between gap-3 border-b border-slate-100/50 pb-2.5 last:border-0 last:pb-0">
+                          <div className="flex items-start gap-2.5">
+                            {msg.sender === "user" ? (
+                              <Man1Avatar size="h-6 w-6" />
+                            ) : (
+                              <RobotAvatar bg="bg-[var(--brand-200)]/50" size="h-6 w-6" />
+                            )}
+                            <div>
+                              <span className={`text-[10.5px] font-bold block mb-0.5 ${msg.sender === 'user' ? 'text-primary' : 'text-slate-700'}`}>
+                                {msg.senderName}
+                              </span>
+                              {/* Pre-text spinner: shown only when streaming AND no content has arrived. */}
+                              {isStreaming && !hasContent ? (
+                                <span className="inline-flex items-center gap-1.5 text-[11px] text-slate-400 italic">
+                                  <Loader2 className="w-3 h-3 animate-spin text-primary" />
+                                  正在思考...
+                                </span>
+                              ) : (
+                                <span className="text-xs text-slate-700 leading-normal font-normal whitespace-pre-wrap break-words">
+                                  {msg.content}
+                                  {isStreaming && (
+                                    <span className="inline-block w-1 h-3 ml-0.5 bg-primary align-middle animate-pulse" />
+                                  )}
+                                </span>
+                              )}
+                            </div>
                           </div>
+                          <span className="text-[9px] text-slate-400 font-mono shrink-0">{msg.time}</span>
                         </div>
-                        <span className="text-[9px] text-slate-400 font-mono shrink-0">{msg.time}</span>
-                      </div>
-                    ))}
+                      );
+                    })}
                   </div>
 
                 </div>
@@ -1662,13 +1775,14 @@ export default function RoundtablePlanningPage() {
                   
                   {/* Left row buttons */}
                   <div className="flex items-center gap-2">
-                    <button 
+                    <button
                       id="send-message-btn"
                       onClick={handleSendMessage}
-                      className="bg-primary text-white hover:bg-[var(--brand-900)] text-xs font-bold px-4 py-2.5 rounded-xl flex items-center gap-1.5 transition-all shadow-[0_2px_8px_rgba(245,34,45,0.15)] active:scale-95"
+                      disabled={isIntentSending || !intentThreadId}
+                      className="bg-primary text-white hover:bg-[var(--brand-900)] disabled:bg-slate-300 disabled:cursor-not-allowed text-xs font-bold px-4 py-2.5 rounded-xl flex items-center gap-1.5 transition-all shadow-[0_2px_8px_rgba(245,34,45,0.15)] active:scale-95"
                     >
-                      <Send className="w-3 h-3" />
-                      发送
+                      {isIntentSending ? <Loader2 className="w-3 h-3 animate-spin" /> : <Send className="w-3 h-3" />}
+                      {isIntentSending ? "发送中" : "发送"}
                     </button>
 
                     <button 
@@ -2526,28 +2640,33 @@ export default function RoundtablePlanningPage() {
                 </div>
                 
                 <div className="p-3 bg-slate-50/50 border border-slate-100 rounded-xl space-y-3 mr-0.5">
-                  <ul className="text-xs text-slate-700 space-y-2.5 font-normal">
-                    <li className="flex items-baseline gap-2 leading-relaxed">
-                      <span className="text-primary text-base leading-none">•</span>
-                      <span><b className="text-slate-900 font-bold">目标</b>: 制定可执行 of 详细规划方案</span>
-                    </li>
-                    <li className="flex items-baseline gap-2 leading-relaxed">
-                      <span className="text-primary text-base leading-none">•</span>
-                      <span><b className="text-slate-900 font-bold">时间</b>: 72小时内完成</span>
-                    </li>
-                    <li className="flex items-baseline gap-2 leading-relaxed">
-                      <span className="text-primary text-base leading-none">•</span>
-                      <span><b className="text-slate-900 font-bold">资源</b>: 预算 ≤ 500万，人员 ≤ 30人</span>
-                    </li>
-                    <li className="flex items-baseline gap-2 leading-relaxed">
-                      <span className="text-primary text-base leading-none">•</span>
-                      <span><b className="text-slate-900 font-bold">优先级</b>: 主方案优先，妥善调配备用</span>
-                    </li>
-                    <li className="flex items-baseline gap-2 leading-relaxed">
-                      <span className="text-primary text-base leading-none">•</span>
-                      <span><b className="text-slate-900 font-bold">待确认</b>: 备选模块范围，资源阈值警戒</span>
-                    </li>
-                  </ul>
+                  {/* Driven by the intent agent's structured summary. Stays in
+                      "等待" state until the agent emits [INTENT_READY]; then
+                      flips to live objective + constraints + assumptions. */}
+                  {intentReady ? (
+                    <ul className="text-xs text-slate-700 space-y-2.5 font-normal">
+                      <li className="flex items-baseline gap-2 leading-relaxed">
+                        <span className="text-primary text-base leading-none">•</span>
+                        <span><b className="text-slate-900 font-bold">目标</b>: {intentReady.objective}</span>
+                      </li>
+                      {intentReady.constraints.map((c, i) => (
+                        <li key={`c-${i}`} className="flex items-baseline gap-2 leading-relaxed">
+                          <span className="text-primary text-base leading-none">•</span>
+                          <span><b className="text-slate-900 font-bold">约束</b>: {c}</span>
+                        </li>
+                      ))}
+                      {(intentReady.assumptions ?? []).map((a, i) => (
+                        <li key={`a-${i}`} className="flex items-baseline gap-2 leading-relaxed">
+                          <span className="text-slate-400 text-base leading-none">•</span>
+                          <span className="text-slate-500"><b className="text-slate-700 font-bold">假设</b>: {a}</span>
+                        </li>
+                      ))}
+                    </ul>
+                  ) : (
+                    <p className="text-xs text-slate-400 italic leading-relaxed">
+                      等待任务理解智能体澄清完成...
+                    </p>
+                  )}
                 </div>
               </div>
 

offline-backend-20260512/backend/app/gateway/app.py

@@ -32,6 +32,7 @@ from app.gateway.routers import (
     auth,
     channels,
     feedback,
+    intent,
     llm_metrics,
     mcp,
     memory,
@@ -864,6 +865,12 @@ This gateway provides custom endpoints for models, MCP configuration, skills, an
     # Multi-agent orchestration (coordinator + sub-agent broadcast/dispatch)
     app.include_router(multi_agent.router)
 
+    # Task-intent agent (Step 1 of the roundtable flow): single built-in
+    # agent runs a 2-5 turn clarification dialogue, then hands a structured
+    # {objective, constraints, assumptions} summary off to the multi-agent
+    # roundtable as its initial intent.
+    app.include_router(intent.router)
+
     @app.get("/health", tags=["health"])
     async def health_check() -> dict:
         """Health check endpoint.

offline-backend-20260512/backend/app/gateway/routers/intent.py

@@ -0,0 +1,167 @@
+"""Task-intent agent endpoints (Step 1 of the roundtable flow).
+
+Lightweight wrapper around the lead-agent runtime that runs a single built-in
+agent (``roundtable-intent``) on its own thread. The agent's job is to do 2-5
+clarifying turns with the user and then emit a structured
+``{objective, constraints, assumptions}`` summary the frontend hands off to
+the multi-agent roundtable as its initial intent.
+
+Same SSE shape as ``multi_agent.py`` (transparent passthrough + appended
+status frame), so the frontend can reuse the same reader pattern.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+import re
+from typing import Any, AsyncIterator
+
+import httpx
+from fastapi import APIRouter, Body, HTTPException, Request
+from fastapi.responses import StreamingResponse
+
+# Reuse the building blocks already in multi_agent.py — keeps both routers
+# identical in connection handling, payload shape, and error mapping.
+from app.gateway.routers.multi_agent import (
+    _auth_headers,
+    _create_thread,
+    _flatten_content,
+    _loopback_base,
+    _parse_last_messages,
+    _run_payload,
+    _stream_upstream,
+)
+
+logger = logging.getLogger(__name__)
+router = APIRouter(prefix="/api/intent", tags=["intent"])
+
+INTENT_AGENT_ID = "roundtable-intent"
+
+# Matches the SOUL-mandated handoff signal: a [INTENT_READY] line followed by
+# a ```json ... ``` block. The model is told to emit this only when it judges
+# the user's task is fully specified.
+_INTENT_READY_PATTERN = re.compile(
+    r"\[INTENT_READY\]\s*```json\s*(\{.*?\})\s*```",
+    re.DOTALL,
+)
+
+
+@router.post("/init")
+async def init_intent(request: Request, payload: dict = Body(...)) -> dict[str, Any]:
+    """Create a fresh thread for an intent-clarification session.
+
+    Returns the stable built-in ``agent_id`` so the frontend doesn't need to
+    know it. Unlike multi-agent ``/init``, no coordinator is built — the
+    intent agent is long-lived and already registered via the legacy
+    agent-sync path on disk.
+    """
+    base = _loopback_base(request)
+    headers = _auth_headers(request)
+    async with httpx.AsyncClient(timeout=None) as client:
+        thread_id = await _create_thread(client, base, headers)
+    return {
+        "status": "success",
+        "agent_id": INTENT_AGENT_ID,
+        "thread_id": thread_id,
+    }
+
+
+@router.post("/stream")
+async def stream_intent(request: Request, payload: dict = Body(...)) -> StreamingResponse:
+    """Drive one turn of the intent dialogue via SSE.
+
+    Body: ``{thread_id, message, model?}``. Streams upstream LangGraph frames
+    through transparently, then appends one status frame:
+
+      {
+        "status": "asking" | "done" | "error",
+        "content": "<final AI text>",
+        "summary": {...} | null   // only present when status == "done"
+      }
+    """
+    thread_id = str(payload.get("thread_id") or "").strip()
+    message = str(payload.get("message") or "").strip()
+    model_name = str(payload.get("model") or "deepseek-chat")
+    if not thread_id:
+        raise HTTPException(status_code=400, detail="thread_id is required")
+    if not message:
+        raise HTTPException(status_code=400, detail="message is required")
+
+    base = _loopback_base(request)
+    headers = _auth_headers(request)
+
+    async def stream_generator() -> AsyncIterator[str]:
+        async with httpx.AsyncClient(timeout=None) as client:
+            try:
+                body = _run_payload(
+                    agent_name=INTENT_AGENT_ID,
+                    model_name=model_name,
+                    thread_id=thread_id,
+                    new_message=message,
+                    # Intent agent shouldn't dispatch, search, write files, or
+                    # view images — its only tool of note is ask_clarification.
+                    excluded_tools=[
+                        "web_search",
+                        "present_files",
+                        "view_image",
+                        "agent_orchestration",
+                    ],
+                    skill_stop_names=[],
+                )
+
+                captured: list[str] = []
+                async for line, done in _stream_upstream(client, base, headers, thread_id, body):
+                    if line is not None:
+                        yield line if line.endswith("\n") else line + "\n"
+                    elif done is not None:
+                        captured = done
+
+                messages = _parse_last_messages(captured)
+                # Find the latest AIMessage's text content — that's what the
+                # frontend renders and what we scan for the handoff signal.
+                last_ai: dict[str, Any] = {}
+                for msg in reversed(messages):
+                    if not isinstance(msg, dict):
+                        continue
+                    if (msg.get("type") or "").lower() in ("ai", "aimessage", "aimessagechunk"):
+                        last_ai = msg
+                        break
+                content = _flatten_content(last_ai.get("content", ""))
+
+                summary: dict[str, Any] | None = None
+                match = _INTENT_READY_PATTERN.search(content)
+                if match:
+                    try:
+                        summary = json.loads(match.group(1))
+                    except json.JSONDecodeError as exc:
+                        logger.warning("intent summary JSON parse failed: %s", exc)
+                        summary = None
+
+                status_value = "done" if summary else "asking"
+                final_frame: dict[str, Any] = {
+                    "status": status_value,
+                    "content": content,
+                    "summary": summary,
+                }
+                yield f"data: {json.dumps(final_frame, ensure_ascii=False)}\n\n"
+
+            except HTTPException as exc:
+                logger.warning("intent stream HTTPException: %s %s", exc.status_code, exc.detail)
+                err = {"status": "error", "content": f"[{exc.status_code}] {exc.detail}", "summary": None}
+                yield f"data: {json.dumps(err, ensure_ascii=False)}\n\n"
+            except Exception as exc:
+                logger.exception("intent stream unexpected failure")
+                err = {"status": "error", "content": f"unexpected: {exc!r}", "summary": None}
+                yield f"data: {json.dumps(err, ensure_ascii=False)}\n\n"
+
+    return StreamingResponse(
+        stream_generator(),
+        media_type="text/event-stream",
+        headers={
+            "Cache-Control": "no-cache",
+            "X-Accel-Buffering": "no",
+            "Connection": "keep-alive",
+        },
+    )