diff --git a/WORKBOOK_v6.md b/WORKBOOK_v6.md
index 9b0ff49..d3acc38 100644
--- a/WORKBOOK_v6.md
+++ b/WORKBOOK_v6.md
@@ -17,17 +17,17 @@
 schema_version: 6
 product: ordinary-user-loop-os
 version_target: loop-os-v1
-current_phase: V6-P2          # V6-P0..P6，见 §3；P0 文档部分 + P1 协议随本 PR 落地
-current_substep: p2_card_cockpit_ui_pending
-last_session_id: s_v6_0001
+current_phase: V6-P3          # V6-P0..P6，见 §3；P2 卡片化座舱随本 PR 落地
+current_substep: p3_real_proof_closeout_pending
+last_session_id: s_v6_0002
 open_holds: 0
 blocked_on: none
-test_baseline: 818            # main 基线，0 fail；本周期任何回归即闸红
+test_baseline: 864            # main 基线，0 fail；本周期任何回归即闸红
 merge_policy: human_only      # 系统永不 merge；auto-merge 本周期禁用
 # next_action 硬上限 2 行：
 next_action: |
-  V6-P0 文档收口 + V6-P1 通信协议（5 卡契约 + daemon 卡片派生）已随本 PR 交付。
-  下一步 V6-P2：座舱 UI 只渲染 5 卡，浏览器 E2E 证明"不读日志即知下一步"。
+  V6-P2 已交付：LoopCard 五卡主表面 + 浏览器 E2E 每步断言卡型与 next_step（证据入库）。
+  下一步 V6-P3：真实证明收口——真 Draft PR、库内真实 Gemini 判词、fail-closed 回归。
 ```
 
 ---
diff --git a/apps/dashboard/src/pages/Cockpit.tsx b/apps/dashboard/src/pages/Cockpit.tsx
index 2053492..a79b516 100644
--- a/apps/dashboard/src/pages/Cockpit.tsx
+++ b/apps/dashboard/src/pages/Cockpit.tsx
@@ -2,17 +2,20 @@ import { useEffect, useMemo, useState, type CSSProperties } from 'react'
 import {
   api,
   ApiError,
+  type ApiLoopCard,
   type ApiMissionOverview,
   type ApiOperatorChoice,
-  type ApiOperatorMissionView,
   type ApiOperatorMessage,
+  type ApiOperatorMissionView,
   type ApiOperatorSession,
   type ApiRepo,
+  type ApiUnderstandingLoopCard,
 } from '../api.js'
 import { useSSE } from '../hooks/useSSE.js'
 import { ChatThread } from './cockpit/ChatThread.js'
 import { Composer } from './cockpit/Composer.js'
 import { ClarificationPopup } from './cockpit/ClarificationPopup.js'
+import { LoopCard } from './cockpit/LoopCard.js'
 import './cockpit/cockpit.css'
 
 const DEFAULT_PROMPT = 'Brainstorm a low-risk improvement, produce PRD/ADR/roadmap, then execute to the draft PR/evidence gate.'
@@ -219,6 +222,14 @@ export function CockpitPage({ onNavigate }: { onNavigate?: (tab: string) => void
   const hasSession = Boolean(session)
   const pendingCount = Math.max(pendingApprovals, sse.pendingApprovals)
   const operatorView = overview?.operatorView
+  // v6-P2 (GR#11) — the primary state surface is exactly ONE of the five loop
+  // cards. The daemon derives it on every mission overview; before a mission
+  // exists (brainstorm/clarify) a client-side understanding card keeps the
+  // same five-card mental model without inventing machine state.
+  const loopCard = useMemo<ApiLoopCard | null>(() => {
+    if (operatorView?.card) return operatorView.card
+    return buildPreMissionUnderstandingCard(session, pendingClarify)
+  }, [operatorView?.card, session, pendingClarify])
   const plannerProvider = operatorView?.providerSummary.planner
   const workerProvider = operatorView?.providerSummary.worker
   const stageForDom = operatorView?.stage ?? (session?.status === 'brainstorming' ? 'brainstorming' : session?.missionId ? 'roadmap_ready' : 'new')
@@ -340,6 +351,8 @@ export function CockpitPage({ onNavigate }: { onNavigate?: (tab: string) => void
           )}
           {notice && <div className="ck-banner notice">{notice}</div>}
 
+          {loopCard && <LoopCard card={loopCard} />}
+
           <ChatThread
             messages={messages}
             busy={busy}
@@ -437,6 +450,49 @@ export function mapErrorToHuman(e: unknown): { kind: 'guidance' | 'error'; text:
   return { kind: 'error', text: GENERIC_HUMAN_ERROR, detail: err.message }
 }
 
+/**
+ * v6-P2 — before a mission/overview exists (brainstorm + clarify rounds) the
+ * daemon has no card to derive, but the ordinary user must still see the same
+ * five-card surface. This builds an honest client-side UnderstandingCard from
+ * the session prompt and the pending clarification questions only; machine
+ * codes stay out of visible text (rule 1), and next_step is always non-empty
+ * (rule 3). Holds keep their dedicated banner — no card is fabricated there.
+ */
+export function buildPreMissionUnderstandingCard(
+  session: ApiOperatorSession | null,
+  pendingClarify: ApiOperatorMessage | null,
+): ApiUnderstandingLoopCard | null {
+  if (!session || session.missionId || session.status === 'hold') return null
+  const pending = pendingClarify?.questions ?? []
+  const questions = pending.map((q) => ({ id: q.id, question: q.question }))
+  const default_assumptions = pending
+    .map((q) => {
+      const recommended = q.options.find((o) => o.recommended)
+      return recommended ? `不回答时默认采用 · If unanswered, the default is "${recommended.label}".` : null
+    })
+    .filter((x): x is string => x !== null)
+  const needsInput = questions.length > 0
+  return {
+    type: 'understanding',
+    title: needsInput ? '需要你补充信息 · Needs your input' : '正在理解你的目标 · Understanding your goal',
+    next_step: needsInput
+      ? '回答下方的待确认问题，AI 才能继续生成方案 · Answer the questions below so the plan can continue.'
+      : '稍等片刻，AI 正在确认理解，随后会给出方案 · Hang on — understanding is being confirmed; a plan comes next.',
+    machine: {
+      user_state: needsInput ? 'needs_more_context' : 'understanding',
+      stage: session.status,
+      hold_code: null,
+      pr_gate_code: null,
+    },
+    user_goal: session.prompt?.trim() || '（目标待补充 · goal not provided yet）',
+    interpreted_goal: 'AI 正在阅读你的目标，还没有改任何东西 · The AI is reading your goal; nothing has been changed yet.',
+    out_of_scope: [],
+    confidence: 0,
+    questions,
+    default_assumptions,
+  }
+}
+
 function isHoldResponse(out: unknown): out is { hold: { code: string; reason: string } } {
   return Boolean(
     out &&
diff --git a/apps/dashboard/src/pages/cockpit/LoopCard.test.tsx b/apps/dashboard/src/pages/cockpit/LoopCard.test.tsx
new file mode 100644
index 0000000..b979b9d
--- /dev/null
+++ b/apps/dashboard/src/pages/cockpit/LoopCard.test.tsx
@@ -0,0 +1,207 @@
+// @vitest-environment jsdom
+/**
+ * v6-P2 — card cockpit: the ordinary user only ever sees the five loop cards
+ * (understanding / plan / progress / blocker / pr_ready) as the primary state
+ * surface. Contract: docs/product/LOOP_COMMUNICATION_PROTOCOL.md (GR#11).
+ *
+ * Pinned invariants:
+ *  - each card type renders with data-card-type and calm bilingual text;
+ *  - every card shows its `next_step` prominently (cockpit-loop-card-next-step);
+ *  - the `machine` sub-object is NEVER rendered as visible text — raw codes
+ *    (HOLD-*, gate codes, stage tokens) live only in data-* attributes.
+ */
+import { describe, it, expect, afterEach } from 'vitest'
+import { render, cleanup } from '@testing-library/react'
+import { LoopCard } from './LoopCard.js'
+import type {
+  ApiBlockerLoopCard,
+  ApiLoopCard,
+  ApiLoopCardMachine,
+  ApiPlanLoopCard,
+  ApiPrReadyLoopCard,
+  ApiProgressLoopCard,
+  ApiUnderstandingLoopCard,
+} from '../../api.js'
+
+afterEach(cleanup)
+
+const MACHINE: ApiLoopCardMachine = {
+  user_state: 'blocked',
+  stage: 'pr_blocked',
+  hold_code: 'HOLD-BUDGET',
+  pr_gate_code: 'REMOTE_WRITES_DISABLED',
+}
+
+function understanding(): ApiUnderstandingLoopCard {
+  return {
+    type: 'understanding',
+    title: '正在理解你的目标 · Understanding your goal',
+    next_step: '回答上面的问题，AI 才能继续生成方案 · Answer the questions above so the plan can continue.',
+    machine: { user_state: 'needs_more_context', stage: 'clarifying', hold_code: null, pr_gate_code: null },
+    user_goal: 'Make onboarding calmer',
+    interpreted_goal: 'AI 正在阅读你的目标 · The AI is reading your goal.',
+    out_of_scope: [],
+    confidence: 62,
+    questions: [{ id: 'q1', question: 'Which slice first?' }],
+    default_assumptions: ['不回答时默认采用 · If unanswered, the default is "Smallest viable slice".'],
+  }
+}
+
+function plan(): ApiPlanLoopCard {
+  return {
+    type: 'plan',
+    title: '等待你的确认 · Waiting for your go-ahead',
+    next_step: '审阅这份方案；你批准后才会开始动手 · Review this plan; work starts only after you approve it.',
+    machine: { user_state: 'waiting_for_approval', stage: 'roadmap_ready', hold_code: null, pr_gate_code: null },
+    objective: 'Improve one onboarding message',
+    phases: ['Understand · 理解需求', 'Execute · 本地执行'],
+    acceptance_criteria: ['gate.json'],
+    risk_level: 'low',
+    estimated_calls: 15,
+    requires_approval: true,
+  }
+}
+
+function progress(): ApiProgressLoopCard {
+  return {
+    type: 'progress',
+    title: '正在执行 · Working on it',
+    next_step: 'Watch progress here; evidence lands automatically · 进度会自动更新。',
+    machine: { user_state: 'executing', stage: 'running', hold_code: null, pr_gate_code: null },
+    current_phase: '正在执行 · Working on it',
+    current_action: 'worker 正在按方案干活 · The worker is following the plan.',
+    evidence_links: ['evidence/run-1/gate.json'],
+    tests_run: ['typecheck', 'vitest'],
+  }
+}
+
+function blocker(): ApiBlockerLoopCard {
+  return {
+    type: 'blocker',
+    title: '需要你处理 · Needs your attention',
+    next_step: '等到明天额度自动恢复 · Wait for the allowance to reset tomorrow.',
+    machine: MACHINE,
+    human_explanation: '今日预算已用完，明天自动恢复或调高预算 · Today’s budget is used up; it resets tomorrow, or you can raise the limit.',
+    why_it_matters: '预算护栏防止系统超额消耗调用额度 · The budget guard stops silent overspend.',
+    recovery_actions: ['等到明天额度自动恢复 · Wait for the reset.', '调高今日预算 · Raise today’s budget.'],
+    recommended_action: '等到明天额度自动恢复 · Wait for the reset.',
+  }
+}
+
+function prReady(withUrl: boolean): ApiPrReadyLoopCard {
+  return {
+    type: 'pr_ready',
+    title: '已完成 · Done',
+    next_step: '去 GitHub 审阅这个 Draft PR；merge 由你亲自点 · Review the draft PR; merging is yours.',
+    machine: { user_state: 'completed', stage: 'pr_created', hold_code: null, pr_gate_code: null },
+    pr_url: withUrl ? 'https://github.com/o/r/pull/7' : null,
+    summary: '本轮工作已完成 · This round of work is complete.',
+    files_changed: ['src/a.ts'],
+    tests: ['vitest'],
+    validator_verdict: null,
+    risk: 'low',
+    merge_policy: '只有你能 merge · Human merge only: the system never merges.',
+    rework_button: { enabled: true, label: '不满意？让 AI 返工 · Not satisfied? Ask for rework.' },
+  }
+}
+
+function renderCard(card: ApiLoopCard) {
+  const { container } = render(<LoopCard card={card} />)
+  const root = container.querySelector('[data-testid="cockpit-loop-card"]') as HTMLElement
+  expect(root).toBeTruthy()
+  return root
+}
+
+describe('LoopCard — five card types render with a prominent next_step', () => {
+  const cases: Array<[string, ApiLoopCard]> = [
+    ['understanding', understanding()],
+    ['plan', plan()],
+    ['progress', progress()],
+    ['blocker', blocker()],
+    ['pr_ready', prReady(false)],
+  ]
+
+  for (const [type, card] of cases) {
+    it(`renders the ${type} card with data-card-type and visible next_step`, () => {
+      const root = renderCard(card)
+      expect(root.getAttribute('data-card-type')).toBe(type)
+      const next = root.querySelector('[data-testid="cockpit-loop-card-next-step"]') as HTMLElement
+      expect(next).toBeTruthy()
+      expect(next.textContent).toContain(card.next_step)
+      expect(root.textContent).toContain(card.title)
+    })
+  }
+})
+
+describe('LoopCard — per-type content', () => {
+  it('understanding: shows goal, interpreted goal, questions and default assumptions', () => {
+    const root = renderCard(understanding())
+    expect(root.textContent).toContain('Make onboarding calmer')
+    expect(root.textContent).toContain('The AI is reading your goal')
+    expect(root.textContent).toContain('Which slice first?')
+    expect(root.textContent).toContain('Smallest viable slice')
+    expect(root.textContent).toContain('62')
+  })
+
+  it('plan: shows objective, phases, acceptance criteria, and the approval requirement', () => {
+    const root = renderCard(plan())
+    expect(root.textContent).toContain('Improve one onboarding message')
+    expect(root.textContent).toContain('Understand · 理解需求')
+    expect(root.textContent).toContain('gate.json')
+    // requires_approval=true must be visible as calm human text
+    expect(root.textContent).toMatch(/批准|approve/i)
+  })
+
+  it('progress: shows current action, checks, and evidence links', () => {
+    const root = renderCard(progress())
+    expect(root.textContent).toContain('The worker is following the plan')
+    expect(root.textContent).toContain('typecheck')
+    expect(root.textContent).toContain('evidence/run-1/gate.json')
+  })
+
+  it('pr_ready: honest about a missing PR url and missing verdict; merge policy always visible', () => {
+    const root = renderCard(prReady(false))
+    expect(root.textContent).toContain('Human merge only')
+    expect(root.textContent).toMatch(/尚未创建|No draft PR/)
+    expect(root.textContent).toMatch(/还没有结论|No review verdict/)
+    expect(root.querySelector('a')).toBeNull()
+  })
+
+  it('pr_ready: renders the PR link when a real url exists', () => {
+    const root = renderCard(prReady(true))
+    const link = root.querySelector('a')
+    expect(link?.getAttribute('href')).toBe('https://github.com/o/r/pull/7')
+  })
+})
+
+describe('LoopCard — blocker card: human explanation, never raw codes', () => {
+  it('shows human_explanation, why_it_matters and recovery actions', () => {
+    const root = renderCard(blocker())
+    expect(root.textContent).toContain('今日预算已用完')
+    expect(root.textContent).toContain('The budget guard stops silent overspend')
+    expect(root.textContent).toContain('Raise today’s budget')
+  })
+
+  it('never renders raw machine codes as visible text; they stay in data-* attributes', () => {
+    const root = renderCard(blocker())
+    expect(root.textContent).not.toContain('HOLD-BUDGET')
+    expect(root.textContent).not.toContain('REMOTE_WRITES_DISABLED')
+    expect(root.textContent).not.toContain('pr_blocked')
+    expect(root.getAttribute('data-hold-code')).toBe('HOLD-BUDGET')
+    expect(root.getAttribute('data-pr-gate-code')).toBe('REMOTE_WRITES_DISABLED')
+    expect(root.getAttribute('data-machine-stage')).toBe('pr_blocked')
+    expect(root.getAttribute('data-user-state')).toBe('blocked')
+  })
+
+  it('keeps machine tokens out of visible text for every card type', () => {
+    const cards: ApiLoopCard[] = [understanding(), plan(), progress(), blocker(), prReady(false)]
+    for (const card of cards) {
+      const root = renderCard(card)
+      for (const token of [card.machine.user_state, card.machine.stage, card.machine.hold_code, card.machine.pr_gate_code]) {
+        if (!token) continue
+        expect(root.textContent).not.toContain(token)
+      }
+      cleanup()
+    }
+  })
+})
diff --git a/apps/dashboard/src/pages/cockpit/LoopCard.tsx b/apps/dashboard/src/pages/cockpit/LoopCard.tsx
new file mode 100644
index 0000000..de26dc5
--- /dev/null
+++ b/apps/dashboard/src/pages/cockpit/LoopCard.tsx
@@ -0,0 +1,172 @@
+/**
+ * v6-P2 — the five-card loop cockpit surface (WORKBOOK_v6 GR#11).
+ *
+ * Renders exactly ONE of the five ordinary-user cards derived by the daemon
+ * (`overview.operatorView.card`, packages/daemon/src/loop-cards.ts). Three
+ * rules from docs/product/LOOP_COMMUNICATION_PROTOCOL.md:
+ *  1. Machine codes stay in the data layer — the `machine` sub-object is
+ *     exposed ONLY through data-* attributes, never as visible text.
+ *  2. Visible text is calm and bilingual.
+ *  3. Every card answers "what happens next" via a prominent `next_step`.
+ */
+import type { ReactNode } from 'react'
+import type { ApiLoopCard } from '../../api.js'
+
+const TYPE_LABELS: Record<ApiLoopCard['type'], string> = {
+  understanding: '理解 · Understanding',
+  plan: '方案 · Plan',
+  progress: '进展 · Progress',
+  blocker: '需要你 · Needs you',
+  pr_ready: '收尾 · Ready for you',
+}
+
+const RISK_LABELS: Record<'low' | 'medium' | 'high', string> = {
+  low: '低 · low',
+  medium: '中 · medium',
+  high: '高 · high',
+}
+
+function Row({ k, children }: { k: string; children: ReactNode }) {
+  return (
+    <div className="ck-loop-row">
+      <dt>{k}</dt>
+      <dd>{children}</dd>
+    </div>
+  )
+}
+
+function List({ items }: { items: string[] }) {
+  return (
+    <ul className="ck-loop-list">
+      {items.map((item, i) => <li key={i}>{item}</li>)}
+    </ul>
+  )
+}
+
+function UnderstandingBody({ card }: { card: Extract<ApiLoopCard, { type: 'understanding' }> }) {
+  return (
+    <dl className="ck-loop-body">
+      <Row k="目标 · Goal">{card.user_goal}</Row>
+      <Row k="理解 · Reading">{card.interpreted_goal}</Row>
+      <Row k="把握 · Confidence">{card.confidence > 0 ? `${card.confidence}%` : '评估中 · being assessed'}</Row>
+      {card.out_of_scope.length > 0 && <Row k="不做 · Out of scope"><List items={card.out_of_scope} /></Row>}
+      {card.questions.length > 0 && (
+        <Row k="待确认 · Questions"><List items={card.questions.map((q) => q.question)} /></Row>
+      )}
+      {card.default_assumptions.length > 0 && (
+        <Row k="默认 · Defaults"><List items={card.default_assumptions} /></Row>
+      )}
+    </dl>
+  )
+}
+
+function PlanBody({ card }: { card: Extract<ApiLoopCard, { type: 'plan' }> }) {
+  return (
+    <dl className="ck-loop-body">
+      <Row k="目标 · Objective">{card.objective}</Row>
+      <Row k="步骤 · Phases">
+        {card.phases.length > 0
+          ? <List items={card.phases} />
+          : '步骤会随方案展开 · Phases appear as the plan takes shape.'}
+      </Row>
+      <Row k="验收 · Acceptance"><List items={card.acceptance_criteria} /></Row>
+      <Row k="风险 · Risk">{RISK_LABELS[card.risk_level]}</Row>
+      <Row k="预算 · Budget">{`最多 ${card.estimated_calls} 次调用 · up to ${card.estimated_calls} calls`}</Row>
+      {card.requires_approval && (
+        <Row k="批准 · Approval">需要你批准后才会动手；merge 永远由你执行 · Nothing runs until you approve; merging stays yours.</Row>
+      )}
+    </dl>
+  )
+}
+
+function ProgressBody({ card }: { card: Extract<ApiLoopCard, { type: 'progress' }> }) {
+  return (
+    <dl className="ck-loop-body">
+      <Row k="阶段 · Phase">{card.current_phase}</Row>
+      <Row k="正在做 · Doing">{card.current_action}</Row>
+      <Row k="检查 · Checks">
+        {card.tests_run.length > 0 ? <List items={card.tests_run} /> : '还没有检查记录 · No checks recorded yet.'}
+      </Row>
+      <Row k="证据 · Evidence">
+        {card.evidence_links.length > 0
+          ? <List items={card.evidence_links} />
+          : '证据生成后会列在这里 · Evidence appears here once produced.'}
+      </Row>
+    </dl>
+  )
+}
+
+function BlockerBody({ card }: { card: Extract<ApiLoopCard, { type: 'blocker' }> }) {
+  return (
+    <>
+      <p className="ck-loop-explanation" data-testid="cockpit-loop-card-explanation">{card.human_explanation}</p>
+      <dl className="ck-loop-body">
+        <Row k="为什么 · Why">{card.why_it_matters}</Row>
+        <Row k="你可以 · You can"><List items={card.recovery_actions} /></Row>
+        <Row k="建议 · Recommended">{card.recommended_action}</Row>
+      </dl>
+    </>
+  )
+}
+
+function PrReadyBody({ card, onRework }: { card: Extract<ApiLoopCard, { type: 'pr_ready' }>; onRework?: (() => void) | undefined }) {
+  return (
+    <>
+      <dl className="ck-loop-body">
+        <Row k="摘要 · Summary">{card.summary}</Row>
+        <Row k="Draft PR">
+          {card.pr_url
+            ? <a href={card.pr_url} target="_blank" rel="noreferrer">{card.pr_url}</a>
+            : '尚未创建真实 PR · No draft PR has been created yet.'}
+        </Row>
+        <Row k="改动 · Changed">
+          {card.files_changed.length > 0 ? <List items={card.files_changed} /> : '没有记录到文件改动 · No file changes recorded.'}
+        </Row>
+        <Row k="检查 · Checks">
+          {card.tests.length > 0 ? <List items={card.tests} /> : '还没有检查记录 · No checks recorded yet.'}
+        </Row>
+        <Row k="评审 · Review">{card.validator_verdict ?? '结果评审还没有结论 · No review verdict yet.'}</Row>
+        <Row k="风险 · Risk">{RISK_LABELS[card.risk]}</Row>
+        <Row k="合并 · Merge">{card.merge_policy}</Row>
+      </dl>
+      <div className="ck-loop-actions">
+        <button
+          type="button"
+          className="ck-btn ck-loop-rework"
+          disabled={!card.rework_button.enabled || !onRework}
+          onClick={onRework}
+        >
+          {card.rework_button.label}
+        </button>
+      </div>
+    </>
+  )
+}
+
+export function LoopCard({ card, onRework }: { card: ApiLoopCard; onRework?: () => void }) {
+  return (
+    <section
+      className={`ck-loop-card type-${card.type}`}
+      data-testid="cockpit-loop-card"
+      data-card-type={card.type}
+      data-user-state={card.machine.user_state}
+      data-machine-stage={card.machine.stage}
+      data-hold-code={card.machine.hold_code ?? ''}
+      data-pr-gate-code={card.machine.pr_gate_code ?? ''}
+    >
+      <div className="ck-loop-head">
+        <span className={`ck-loop-type type-${card.type}`}>{TYPE_LABELS[card.type]}</span>
+        <strong className="ck-loop-title">{card.title}</strong>
+      </div>
+      <div className="ck-loop-next" data-testid="cockpit-loop-card-next-step">
+        <span className="k">下一步 · Next</span>
+        <span className="v">{card.next_step}</span>
+      </div>
+      {card.type === 'understanding' && <UnderstandingBody card={card} />}
+      {card.type === 'plan' && <PlanBody card={card} />}
+      {card.type === 'progress' && <ProgressBody card={card} />}
+      {card.type === 'blocker' && <BlockerBody card={card} />}
+      {card.type === 'pr_ready' && <PrReadyBody card={card} onRework={onRework} />}
+    </section>
+  )
+}
diff --git a/apps/dashboard/src/pages/cockpit/cockpit.css b/apps/dashboard/src/pages/cockpit/cockpit.css
index 8e69bf9..ddf7415 100644
--- a/apps/dashboard/src/pages/cockpit/cockpit.css
+++ b/apps/dashboard/src/pages/cockpit/cockpit.css
@@ -280,6 +280,27 @@ button.ck-stat:hover { background: var(--bg-sub); }
 .ck-validator-top { display: flex; align-items: center; justify-content: space-between; gap: 8px; font-size: 12px; }
 .ck-validator-top span { font-size: 10.5px; text-transform: uppercase; letter-spacing: 0.04em; color: var(--fg-muted); }
 
+/* --- v6-P2 five-card loop surface (GR#11): the primary state card above the chat --- */
+.ck-loop-card { border: 1px solid #dbe4f5; border-radius: var(--r-lg); background: var(--bg); box-shadow: var(--sh-1); padding: 12px 14px; margin-bottom: 8px; display: grid; gap: 8px; font-size: 12.5px; }
+.ck-loop-card.type-blocker { border-color: #fed7aa; background: #fffbf5; }
+.ck-loop-card.type-pr_ready { border-color: #a7f3d0; background: #f6fdf9; }
+.ck-loop-head { display: flex; align-items: center; gap: 10px; min-width: 0; }
+.ck-loop-type { flex: none; border-radius: 999px; padding: 2px 9px; font-size: 10.5px; font-weight: 600; text-transform: none; background: var(--accent-soft); color: #1d4ed8; }
+.ck-loop-type.type-blocker { background: #fff1e6; color: #c2410c; }
+.ck-loop-type.type-pr_ready { background: #ecfdf5; color: #047857; }
+.ck-loop-title { font-size: 13.5px; overflow: hidden; text-overflow: ellipsis; }
+.ck-loop-next { display: flex; align-items: baseline; gap: 8px; border: 1px solid #bfdbfe; background: var(--accent-soft); border-radius: 10px; padding: 8px 10px; }
+.ck-loop-next .k { flex: none; font-size: 10.5px; font-weight: 700; letter-spacing: 0.03em; color: #1d4ed8; }
+.ck-loop-next .v { font-size: 13px; font-weight: 600; color: var(--fg); }
+.ck-loop-explanation { margin: 0; font-size: 13px; color: var(--fg-2); line-height: 1.5; }
+.ck-loop-body { margin: 0; display: grid; gap: 4px; }
+.ck-loop-row { display: grid; grid-template-columns: 110px minmax(0, 1fr); gap: 8px; }
+.ck-loop-row dt { color: var(--fg-muted); }
+.ck-loop-row dd { margin: 0; color: var(--fg-2); overflow-wrap: anywhere; }
+.ck-loop-list { margin: 0; padding-left: 18px; display: grid; gap: 2px; }
+.ck-loop-actions { display: flex; gap: 8px; }
+.ck-loop-rework { font-size: 12px; }
+
 .ck-review-block { align-self: stretch; max-width: 92%; border: var(--hair); border-radius: var(--r-lg); background: var(--bg); box-shadow: var(--sh-1); padding: 12px; display: grid; gap: 10px; }
 .ck-review-head { display: flex; align-items: center; justify-content: space-between; gap: 8px; font-size: 13px; }
 .ck-review-head span { color: var(--fg-muted); font-size: 11px; text-transform: uppercase; letter-spacing: 0.04em; }
diff --git a/docs/SESSION_LOG_v3.md b/docs/SESSION_LOG_v3.md
index a51373e..a1c5cea 100644
--- a/docs/SESSION_LOG_v3.md
+++ b/docs/SESSION_LOG_v3.md
@@ -1,5 +1,14 @@
 # SESSION LOG v3
 
+## s_v6_0002 · 2026-06-11 · V6-P2 complete · card cockpit (UI renders the five loop cards)
+
+- New `apps/dashboard/src/pages/cockpit/LoopCard.tsx` (TDD: 13 component tests first): renders exactly the five GR#11 card types from `overview.operatorView.card`; calm bilingual copy; `next_step` is always the prominent first row (`cockpit-loop-card-next-step`); the `machine` sub-object is never visible text — raw codes live only in `data-card-type` / `data-user-state` / `data-machine-stage` / `data-hold-code` / `data-pr-gate-code`. Blocker card shows `human_explanation` + `why_it_matters` + recovery actions, zero raw codes.
+- Mounted in `Cockpit.tsx` above the chat as the primary state surface. Before a mission exists (brainstorm/clarify), an honest client-side UnderstandingCard keeps the same five-card mental model (questions + recommended defaults from the pending clarification; no fabricated machine state; session holds keep their dedicated banner). All pre-existing testids unchanged.
+- `scripts/operator-cockpit-user-e2e.ts` extended: every journey step now also asserts the loop card is present with the EXPECTED type — understanding during clarify (steps 2–3), plan at roadmap_ready (step 4), progress during execution (step 5), blocker at the Draft PR gate (step 7) — plus non-empty `next_step` visible without logs and no machine token inside the card's visible text. All previous assertions kept.
+- Browser evidence (real chromium, mock/template env — simulated planner/worker, real UI): user-journey E2E PASS 7/7 at `evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/`; quality smoke PASS at `evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/`.
+- Gates: `pnpm typecheck` PASS; `pnpm lint` PASS; `pnpm test` PASS with 864 passed, 6 skipped (128 files; +13 over the 851 baseline, zero regressions).
+- §0 → `current_phase: V6-P3`（真实证明收口）。L1 of V6-P2 met: "不读日志即知下一步" now has browser-level evidence; the old loop summary stays as a read-only evidence detail in the thread footer, no longer the primary state surface.
+
 ## s_0013 · 2026-06-10 · operator-reported real-chain regression PASS (closed-gate) · P4 still pending open-gate run
 
 - Operator ran a real Mac E2E against `hermus-agent` (temp clone): planner=`claude-cli/local_claude_code`, coder=`codex-cli/local_codex`, **Gemini verdict `pass`** persisted to `validator-summary.json`, Draft-PR gate blocked with `REMOTE_WRITES_DISABLED`; no push, no PR URL, no merge; hermus-agent itself untouched. Report (operator machine): `evidence/launch/operator-cockpit-real-smoke-2026-06-10T07-06-36-411Z.md`. Operator gates green locally.
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/01-new.png b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/01-new.png
new file mode 100644
index 0000000..cbce33e
Binary files /dev/null and b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/01-new.png differ
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/02-brainstorm-ready.png b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/02-brainstorm-ready.png
new file mode 100644
index 0000000..1611f69
Binary files /dev/null and b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/02-brainstorm-ready.png differ
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/03-plan-approval.png b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/03-plan-approval.png
new file mode 100644
index 0000000..0b838bb
Binary files /dev/null and b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/03-plan-approval.png differ
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/04-approved.png b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/04-approved.png
new file mode 100644
index 0000000..fbde3f9
Binary files /dev/null and b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/04-approved.png differ
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/05-evidence-ready.png b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/05-evidence-ready.png
new file mode 100644
index 0000000..b2fc80d
Binary files /dev/null and b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/05-evidence-ready.png differ
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/06-pr-blocked.png b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/06-pr-blocked.png
new file mode 100644
index 0000000..3281adf
Binary files /dev/null and b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/06-pr-blocked.png differ
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/console-logs.json b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/console-logs.json
new file mode 100644
index 0000000..0637a08
--- /dev/null
+++ b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/console-logs.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/db-state-summary.json b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/db-state-summary.json
new file mode 100644
index 0000000..23cba9a
--- /dev/null
+++ b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/db-state-summary.json
@@ -0,0 +1,286 @@
+{
+  "mission": {
+    "id": "01KTT64DP0JWA5A7A9R0M0VMXJ",
+    "status": "paused",
+    "githubPrUrl": null
+  },
+  "operatorView": {
+    "stage": "pr_blocked",
+    "stageLabel": "PR blocked by policy · PR 被安全门拦截",
+    "confidence": 96,
+    "progressPercent": 95,
+    "headlessCallsToday": 0,
+    "primaryAction": {
+      "id": "check-draft-pr-gate",
+      "label": "Re-check Draft PR Gate · 重新检查 PR 安全门",
+      "kind": "primary"
+    },
+    "secondaryActions": [],
+    "providerSummary": {
+      "planner": {
+        "name": "test-synthetic",
+        "mode": "mock",
+        "status": "Planner finished",
+        "tokens": null
+      },
+      "worker": {
+        "name": "mock",
+        "mode": "mock",
+        "status": "done",
+        "tokens": null
+      },
+      "validators": [
+        {
+          "name": "gemini",
+          "mode": "not_configured",
+          "status": "not_configured"
+        }
+      ]
+    },
+    "safetySummary": {
+      "remoteWrites": "disabled",
+      "prGate": {
+        "status": "blocked",
+        "code": "GEMINI_NOT_CONFIGURED",
+        "reason": "Gemini hard gate has no evidence-only PASS verdict for this mission.",
+        "remediation": "Remote writes are disabled for safety. Enable repo-scoped allow_remote_writes only when you want the worker to push a branch and open a Draft PR; until then no push, PR, or merge occurs."
+      },
+      "testMode": {
+        "enabled": true,
+        "reason": "mock/template mode is active; no external model or remote write is implied."
+      }
+    },
+    "understanding": {
+      "roundsCompleted": 0,
+      "questions": [],
+      "readyReason": "Planner confidence is at least 95% and no clarification questions are pending."
+    },
+    "projectPulse": {
+      "progress": [
+        {
+          "id": "understand",
+          "label": "Understand · 理解需求",
+          "status": "done"
+        },
+        {
+          "id": "roadmap",
+          "label": "Roadmap · 路线图",
+          "status": "done"
+        },
+        {
+          "id": "execute",
+          "label": "Execute · 本地执行",
+          "status": "done"
+        },
+        {
+          "id": "validate",
+          "label": "Validate · 独立验证",
+          "status": "done",
+          "detail": "Gemini key is not configured; this is visible and not counted as pass."
+        },
+        {
+          "id": "pr-gate",
+          "label": "PR Gate · PR 安全门",
+          "status": "active"
+        },
+        {
+          "id": "learn",
+          "label": "Learn · 沉淀记忆",
+          "status": "pending"
+        }
+      ],
+      "workingFolder": "/tmp/aedev-cockpit-quality-BEYx80/operator-evidence/01KTT64E24GGCQ2YWX6V7VYQCW",
+      "touchedFiles": [],
+      "evidence": [
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFK",
+          "title": "ADR draft",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/adr-mission.md",
+          "type": "adr"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFH",
+          "title": "Evidence directory",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ",
+          "type": "evidence"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFJ",
+          "title": "PRD",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/prd.md",
+          "type": "prd"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFN",
+          "title": "Workbook summary",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/workbook-summary.md",
+          "type": "report"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFP",
+          "title": "Test summary",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/test-summary.md",
+          "type": "report"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFQ",
+          "title": "Risk report",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/risk-report.md",
+          "type": "report"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFR",
+          "title": "Worker diff",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/diff-summary.md",
+          "type": "report"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFS",
+          "title": "Done report",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/done-report.md",
+          "type": "report"
+        },
+        {
+          "id": "01KTT64E2AT9MJC8JFV73ZEVFM",
+          "title": "Roadmap",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ/roadmap.md",
+          "type": "roadmap"
+        },
+        {
+          "id": "01KTT64DP4CEC3RWAFR03NMXJ9",
+          "title": "ADR draft in mission design",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/prd/01KTT64DP0JWA5A7A9R0M0VMXJ.design.json",
+          "type": "adr"
+        },
+        {
+          "id": "01KTT64DP4CEC3RWAFR03NMXJ7",
+          "title": "PRD",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/prd/01KTT64DP0JWA5A7A9R0M0VMXJ.md",
+          "type": "prd"
+        },
+        {
+          "id": "01KTT64DP4CEC3RWAFR03NMXJ8",
+          "title": "Mission design JSON",
+          "path": "/tmp/aedev-cockpit-quality-BEYx80/prd/01KTT64DP0JWA5A7A9R0M0VMXJ.design.json",
+          "type": "roadmap"
+        }
+      ],
+      "validatorReviews": [
+        {
+          "id": "validators-not-configured",
+          "validator": "validators",
+          "verdict": "not_configured",
+          "summary": "Independent validation did not run because the Gemini key is not configured.",
+          "checkedEvidence": [
+            "ADR draft",
+            "Evidence directory",
+            "PRD",
+            "Workbook summary",
+            "Test summary",
+            "Risk report",
+            "Worker diff",
+            "Done report"
+          ],
+          "blockingIssues": [],
+          "evidenceGaps": [
+            "No Gemini validator verdict exists for this mission."
+          ],
+          "recommendedNextAction": "Configure validator keys for live verification, or continue reviewing evidence manually."
+        }
+      ]
+    },
+    "memorySummary": {
+      "projectFacts": [
+        {
+          "id": "repo-01KTT64A17VBM5AED1ERZ9TK1K",
+          "kind": "project",
+          "text": "Target repo is cockpit-quality at /tmp/aedev-cockpit-quality-BEYx80.",
+          "provenance": "repo registry",
+          "ttlDays": 90,
+          "superseded": false
+        },
+        {
+          "id": "repo-forbidden-01KTT64A17VBM5AED1ERZ9TK1K",
+          "kind": "safety",
+          "text": "Forbidden paths stay protected: .env*, secrets/**, .github/**, AGENTS.md",
+          "provenance": "repo policy",
+          "ttlDays": 365,
+          "superseded": false
+        }
+      ],
+      "userPreferences": [
+        {
+          "id": "pref-understand-first",
+          "kind": "user_preference",
+          "text": "Ask goal-specific questions and confirm understanding before starting worker execution.",
+          "provenance": "operator product directive",
+          "ttlDays": 365,
+          "superseded": false
+        },
+        {
+          "id": "prompt-01KTT64B82AYBJ6M976ZHGPSAW",
+          "kind": "mission_intent",
+          "text": "Current mission intent: In the dashboard Cockpit page, verify the existing conversation UI quality smoke keeps the single conversation layout, status strip, and safe Draft PR gate visible without changing product behavior. Acceptance: browser smoke passes and evid",
+          "provenance": "operator prompt",
+          "ttlDays": 30,
+          "superseded": false
+        }
+      ],
+      "recentLessons": [
+        {
+          "id": "lesson-0",
+          "kind": "run_lesson",
+          "text": "Draft PR blocked: GEMINI_NOT_CONFIGURED",
+          "provenance": "event:operator.draft_pr_blocked",
+          "ttlDays": 30,
+          "superseded": false
+        }
+      ]
+    },
+    "summary": "Worker done, evidence ready, and the Draft PR gate was blocked by policy. No branch push, PR, or merge occurred.",
+    "nextAction": "Continue reviewing evidence, or explicitly enable repo-scoped remote writes before re-checking the gate.",
+    "testMode": true,
+    "userState": {
+      "state": "blocked",
+      "label": "Needs your attention",
+      "labelZh": "需要你处理",
+      "explanation": "系统在这一步暂停，等你看一眼后再继续 · The system paused here and will continue once you take a look."
+    },
+    "lastActivity": {
+      "atIso": "2026-06-11T01:53:18.134Z",
+      "agoMs": 159,
+      "phase": "blocked"
+    },
+    "loopSummary": {
+      "whatChanged": [],
+      "testsRan": [
+        "Test summary"
+      ],
+      "agents": [
+        "planner · test-synthetic",
+        "worker · mock",
+        "validator · gemini"
+      ],
+      "validatorSaid": null,
+      "whyStoppedOrContinuing": "系统在这一步暂停，等你看一眼后再继续 · The system paused here and will continue once you take a look."
+    },
+    "card": {
+      "type": "blocker",
+      "title": "需要你处理 · Needs your attention",
+      "human_explanation": "系统在这一步暂停，等你看一眼后再继续 · The system paused here and will continue once you take a look.",
+      "why_it_matters": "在不确定的时候暂停，比悄悄做错更安全；没有你的确认，任何东西都不会对外发布 · Pausing when unsure is safer than quietly doing the wrong thing; nothing is published without your confirmation.",
+      "recovery_actions": [
+        "查看这张卡的说明，确认是否继续 · Read this card’s explanation and confirm whether to continue.",
+        "随时可以重新开始或调整目标 · You can restart or adjust the goal at any time."
+      ],
+      "recommended_action": "查看这张卡的说明，确认是否继续 · Read this card’s explanation and confirm whether to continue.",
+      "next_step": "查看这张卡的说明，确认是否继续 · Read this card’s explanation and confirm whether to continue.",
+      "machine": {
+        "user_state": "blocked",
+        "stage": "pr_blocked",
+        "hold_code": null,
+        "pr_gate_code": "GEMINI_NOT_CONFIGURED"
+      }
+    }
+  }
+}
\ No newline at end of file
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/dom-state-summary.json b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/dom-state-summary.json
new file mode 100644
index 0000000..6e84ddc
--- /dev/null
+++ b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/dom-state-summary.json
@@ -0,0 +1,6 @@
+{
+  "stage": "pr_blocked",
+  "planner": "mock",
+  "worker": "mock",
+  "prGateCode": "GEMINI_NOT_CONFIGURED"
+}
\ No newline at end of file
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/event-tail.json b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/event-tail.json
new file mode 100644
index 0000000..9d6bd56
--- /dev/null
+++ b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/event-tail.json
@@ -0,0 +1,119 @@
+[
+  {
+    "type": "operator.draft_pr_blocked",
+    "payload": {
+      "code": "GEMINI_NOT_CONFIGURED",
+      "reason": "Gemini hard gate has no evidence-only PASS verdict for this mission.",
+      "validator": "gemini"
+    },
+    "createdAt": "2026-06-11T01:53:18.134Z"
+  },
+  {
+    "type": "operator.gemini_pr_blocked",
+    "payload": {
+      "code": "GEMINI_NOT_CONFIGURED",
+      "reason": "Gemini hard gate has no evidence-only PASS verdict for this mission.",
+      "verdict": "not_configured",
+      "summary": null
+    },
+    "createdAt": "2026-06-11T01:53:18.134Z"
+  },
+  {
+    "type": "operator.evidence_written",
+    "payload": {
+      "sessionId": "01KTT64B82AYBJ6M976ZHGPSAW",
+      "evidenceDir": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ"
+    },
+    "createdAt": "2026-06-11T01:53:16.362Z"
+  },
+  {
+    "type": "operator.stage_changed",
+    "payload": {
+      "stage": "PR/Waiting/Blocked",
+      "sessionId": "01KTT64B82AYBJ6M976ZHGPSAW",
+      "status": "waiting"
+    },
+    "createdAt": "2026-06-11T01:53:16.362Z"
+  },
+  {
+    "type": "mission.run_completed",
+    "payload": {
+      "taskId": "01KTT64E24GGCQ2YWX6V7VYQCW",
+      "runId": "01KTT64E24GGCQ2YWX6V7VYQCY",
+      "exitCode": 0,
+      "status": "waiting",
+      "decision": "WAITING",
+      "riskScore": 0,
+      "validatorCount": 0,
+      "releaseDeployUrl": null,
+      "releaseReverted": false,
+      "draftPrUrl": null,
+      "draftPrNumber": null
+    },
+    "createdAt": "2026-06-11T01:53:16.361Z"
+  },
+  {
+    "type": "operator.worker_started",
+    "payload": {
+      "taskId": "01KTT64E24GGCQ2YWX6V7VYQCW",
+      "runId": "01KTT64E24GGCQ2YWX6V7VYQCY",
+      "provider": "mock",
+      "evidenceDir": "/tmp/aedev-cockpit-quality-BEYx80/operator-evidence/01KTT64E24GGCQ2YWX6V7VYQCW"
+    },
+    "createdAt": "2026-06-11T01:53:16.357Z"
+  },
+  {
+    "type": "operator.worker_log",
+    "payload": {
+      "taskId": "01KTT64E24GGCQ2YWX6V7VYQCW",
+      "runId": "01KTT64E24GGCQ2YWX6V7VYQCY",
+      "stream": "stdout",
+      "chunk": "mock worker completed evidence gate"
+    },
+    "createdAt": "2026-06-11T01:53:16.357Z"
+  },
+  {
+    "type": "mission.route_selected",
+    "payload": {
+      "role": "coder",
+      "provider": "mock",
+      "sessionId": null,
+      "concurrency": 1,
+      "holdCode": null,
+      "reason": "worker router not configured"
+    },
+    "createdAt": "2026-06-11T01:53:16.356Z"
+  },
+  {
+    "type": "mission.run_started",
+    "payload": {
+      "evidenceDir": "/tmp/aedev-cockpit-quality-BEYx80/evidence/01KTT64DP0JWA5A7A9R0M0VMXJ"
+    },
+    "createdAt": "2026-06-11T01:53:16.352Z"
+  },
+  {
+    "type": "operator.validators_not_configured",
+    "payload": {
+      "status": "not_configured",
+      "note": "Gemini validator key is not configured; Draft PR remains blocked until Gemini returns PASS."
+    },
+    "createdAt": "2026-06-11T01:53:16.350Z"
+  },
+  {
+    "type": "operator.worker_assigned",
+    "payload": {
+      "sessionId": "01KTT64B82AYBJ6M976ZHGPSAW",
+      "mode": "mock",
+      "availableSessions": 0,
+      "paidApiKeysStripped": true
+    },
+    "createdAt": "2026-06-11T01:53:16.350Z"
+  },
+  {
+    "type": "operator.run_starting",
+    "payload": {
+      "sessionId": "01KTT64B82AYBJ6M976ZHGPSAW"
+    },
+    "createdAt": "2026-06-11T01:53:16.349Z"
+  }
+]
\ No newline at end of file
diff --git a/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/quality-smoke.md b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/quality-smoke.md
new file mode 100644
index 0000000..2050df9
--- /dev/null
+++ b/evidence/browser-cockpit-quality/2026-06-11T01-53-12-152Z/quality-smoke.md
@@ -0,0 +1,16 @@
+# Operator Cockpit WebUI Quality Smoke
+
+Result: PASS
+Mission: 01KTT64DP0JWA5A7A9R0M0VMXJ
+Stage: pr_blocked
+PR gate: GEMINI_NOT_CONFIGURED
+
+Assertions:
+- cockpit renders as one conversation column plus the three-part status strip
+- legacy Project Pulse, sidebar, inspector, and tabbed panels are absent
+- one primary action per stage
+- stable testids for core controls
+- planner/worker provider badges expose mock test mode
+- PR URL stayed empty while Gemini hard gate was not configured
+- draft PR blocked card reassures no push, PR, or merge occurred
+- browser console had no error/warning
\ No newline at end of file
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/01-composed-and-started.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/01-composed-and-started.png
new file mode 100644
index 0000000..33a692a
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/01-composed-and-started.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/02-planning-progress.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/02-planning-progress.png
new file mode 100644
index 0000000..3529e5b
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/02-planning-progress.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03a-clarify-popup-filled.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03a-clarify-popup-filled.png
new file mode 100644
index 0000000..945749e
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03a-clarify-popup-filled.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03b-clarify-answered-gate-guidance.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03b-clarify-answered-gate-guidance.png
new file mode 100644
index 0000000..552e14e
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03b-clarify-answered-gate-guidance.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03c-clarify-unlocked.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03c-clarify-unlocked.png
new file mode 100644
index 0000000..7f1014c
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03c-clarify-unlocked.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/04-roadmap-ready.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/04-roadmap-ready.png
new file mode 100644
index 0000000..669e32a
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/04-roadmap-ready.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05a-approved.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05a-approved.png
new file mode 100644
index 0000000..560816d
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05a-approved.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05b-execution-evidence-gate.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05b-execution-evidence-gate.png
new file mode 100644
index 0000000..9a2da65
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05b-execution-evidence-gate.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/06-loop-summary.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/06-loop-summary.png
new file mode 100644
index 0000000..9a2da65
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/06-loop-summary.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/07-pr-gate-blocked-human.png b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/07-pr-gate-blocked-human.png
new file mode 100644
index 0000000..d22c607
Binary files /dev/null and b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/07-pr-gate-blocked-human.png differ
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/console-logs.json b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/console-logs.json
new file mode 100644
index 0000000..2936d7c
--- /dev/null
+++ b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/console-logs.json
@@ -0,0 +1,3 @@
+[
+  "error: Failed to load resource: the server responded with a status of 409 (Conflict)"
+]
\ No newline at end of file
diff --git a/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/user-e2e-report.md b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/user-e2e-report.md
new file mode 100644
index 0000000..923d8dc
--- /dev/null
+++ b/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/user-e2e-report.md
@@ -0,0 +1,77 @@
+# Operator Cockpit — User Journey E2E Report
+
+Result: **PASS**
+Timestamp: 2026-06-11T01-52-49-157Z
+Evidence dir: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z
+
+Harness: mock/template planner+worker, remote writes disabled, all external CLIs/APIs disabled,
+temp stateDir, in-memory SQLite, vite dashboard, chromium via playwright.
+
+## Steps
+
+### step-1-compose-and-start — PASS
+
+Type a user prompt into the composer and start brainstorm
+- composer testid: cockpit-goal-input · prompt: Make the onboarding flow friendlier for new users. I want it…
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/01-composed-and-started.png
+
+### step-2-visible-progress — PASS
+
+Planning shows visible progress; the UI never looks frozen
+- status strip during planning: STAGE Brainstorm · 共创中 NOW Planner is thinking · Planner 正在分析 PROGRESS 0% — — APPROVALS 0
+- loop card during planning/clarify: type=understanding · next_step="回答下方的待确认问题，AI 才能继续生成方案 · Answer the questions below so the plan can continue."
+- strip refreshed: "STAGE Brainstorm · 共创中 NOW Planner is thinking · Planner 正在分析 PROGRESS 0% — — APPROVALS 0" → "STAGE Decision · 做选择 NOW Review the questions, then generate the plan · 先确认问题，再生成方案 PROGRESS 0% — — APPROVALS 0"
+- cockpit-last-activity refresh check is completed as soon as the mission overview exists (see step 4 notes) — the testid only renders once a mission is created.
+- cockpit-last-activity refresh verified: "LAST ACTIVITY 0s ago" → "LAST ACTIVITY 1s ago" (1.7s apart)
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/02-planning-progress.png
+
+### step-3-clarifications — PASS
+
+Answer the clarification popup through the real UI controls
+- clarification questions rendered: 2
+- answered transcript message visible; popup dismissed
+- locked Generate Plan produced calm guidance, no raw gate code in visible text
+- follow-up round confirmed confidence ≥95; plan unlocked
+- loop card after clarification answers: type=understanding · next_step="稍等片刻，AI 正在确认理解，随后会给出方案 · Hang on — understanding is being confirmed; a plan comes next."
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03a-clarify-popup-filled.png
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03b-clarify-answered-gate-guidance.png
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/03c-clarify-unlocked.png
+
+### step-4-generate-roadmap — PASS
+
+Generate roadmap; PRD/roadmap artifacts exist and stage advances
+- mission 01KTT63TGG0HGHNR7BVJEE704G created with 3 design artifacts (adr, prd, roadmap…)
+- loop card at roadmap_ready: type=plan · next_step="审阅这份方案；你批准后才会开始动手 · Review this plan; work starts only after you approve it."
+- cockpit-last-activity refresh verified: "LAST ACTIVITY 0s ago" → "LAST ACTIVITY 1s ago"
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/04-roadmap-ready.png
+
+### step-5-approve-and-execute — PASS
+
+Approve roadmap, start execution; execution state appears
+- execution state appeared (stage=running)
+- loop card during execution: type=progress · next_step="Wait for progress, pause, or stop if the run is wrong."
+- worker runs recorded: 1; final stage=validators_missing
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05a-approved.png
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/05b-execution-evidence-gate.png
+
+### step-6-loop-summary — PASS
+
+cockpit-loop-summary renders with non-empty whyStoppedOrContinuing
+- whyStoppedOrContinuing: 结果评审尚未配置 · result review not configured
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/06-loop-summary.png
+
+### step-7-draft-pr-gate — PASS
+
+Draft PR gate BLOCKED is calm human text; no raw codes visible
+- machine code stays in data-* only: data-pr-gate-code=GEMINI_NOT_CONFIGURED
+- calm safety phrasing visible (安全门 / no push, no PR, no merge reassurance)
+- loop card at the Draft PR gate: type=blocker · next_step="查看这张卡的说明，确认是否继续 · Read this card’s explanation and confirm whether to continue."
+- no PR URL recorded; operator.draft_pr_blocked event present
+- screenshot: /home/user/claude-code-247/evidence/browser-cockpit-user-e2e/2026-06-11T01-52-49-157Z/07-pr-gate-blocked-human.png
+
+## Browser console issues (informational)
+
+- error: Failed to load resource: the server responded with a status of 409 (Conflict)
+
+> Note: the deliberate locked Generate Plan probe in step 3 produces one expected 409 network log entry;
+> the assertion is that the VISIBLE UI stays human (guidance text, no raw codes).
diff --git a/scripts/operator-cockpit-user-e2e.ts b/scripts/operator-cockpit-user-e2e.ts
index 3908762..aacb77a 100644
--- a/scripts/operator-cockpit-user-e2e.ts
+++ b/scripts/operator-cockpit-user-e2e.ts
@@ -104,6 +104,8 @@ const PLANNED_STEPS: Array<[string, string]> = [
   ['step-6-loop-summary', 'cockpit-loop-summary renders with non-empty whyStoppedOrContinuing'],
   ['step-7-draft-pr-gate', 'Draft PR gate BLOCKED is calm human text; no raw codes visible'],
 ]
+// v6-P2 — at every journey step the five-card loop surface (GR#11) must show
+// the EXPECTED card type with a non-empty, log-free next_step in visible text.
 const steps: StepResult[] = []
 let current: StepResult | undefined
 let harnessError: string | undefined
@@ -208,6 +210,9 @@ async function runJourney(page: Page): Promise<void> {
     // …and WITHOUT any further clicks the planner's brainstorm result must
     // arrive in the conversation (polling keeps the UI alive, not frozen).
     await page.getByText('Initial brainstorm (user e2e fixture)', { exact: false }).first().waitFor({ timeout: 20_000 })
+    // v6-P2: while the system is still understanding/clarifying, the loop card
+    // is the understanding card and its next_step is readable without logs.
+    await expectLoopCard(page, ['understanding'], 'during planning/clarify')
     const stripB = await page.getByTestId('cockpit-status-strip').innerText()
     if (stripA === stripB) {
       // The strip may legitimately settle fast; the brainstorm arrival above
@@ -256,6 +261,8 @@ async function runJourney(page: Page): Promise<void> {
     await page.getByRole('button', { name: /Ask Until Clear/ }).click()
     await page.getByText('your answers are enough to plan safely', { exact: false }).first().waitFor({ timeout: 20_000 })
     note('follow-up round confirmed confidence ≥95; plan unlocked')
+    // v6-P2: still pre-plan — the loop card stays the understanding card.
+    await expectLoopCard(page, ['understanding'], 'after clarification answers')
     await shot(page, '03c-clarify-unlocked')
   })
 
@@ -270,6 +277,8 @@ async function runJourney(page: Page): Promise<void> {
     if (artifacts.length === 0) throw new Error('No roadmap/PRD artifacts registered for the mission')
     note(`mission ${mission.id} created with ${artifacts.length} design artifacts (${artifacts.slice(0, 3).map((a) => a.type).join(', ')}…)`)
     await expectStage(page, ['roadmap_ready', 'pending_approval'])
+    // v6-P2: at roadmap_ready the daemon-derived card is the plan card.
+    await expectLoopCard(page, ['plan'], 'at roadmap_ready')
 
     // Deferred half of step 2: the overview now exists, so cockpit-last-activity
     // must render and keep refreshing (two samples >1.5s apart).
@@ -295,6 +304,8 @@ async function runJourney(page: Page): Promise<void> {
     await page.getByTestId('cockpit-start-execution').click()
     await waitForRootStage(page, ['running', 'evidence_ready', 'validators_missing', 'validating', 'pr_ready'], 30_000)
     note(`execution state appeared (stage=${await rootStage(page)})`)
+    // v6-P2: during execution/checking the loop card is the progress card.
+    await expectLoopCard(page, ['progress'], 'during execution')
     await waitForRootStage(page, ['evidence_ready', 'validators_missing', 'validators_ready', 'pr_ready'], 60_000)
     const mission = db.listMissions()[0]
     if (!mission) throw new Error('Mission disappeared')
@@ -336,6 +347,9 @@ async function runJourney(page: Page): Promise<void> {
     }
     note('calm safety phrasing visible (安全门 / no push, no PR, no merge reassurance)')
     await assertNoForbiddenVisibleText(page, 'on the blocked Draft PR gate screen')
+    // v6-P2: at the gate the loop card is blocker (or pr_ready when a real PR
+    // exists); with remote writes disabled the safe expectation is blocker.
+    await expectLoopCard(page, ['blocker', 'pr_ready'], 'at the Draft PR gate')
     const mission = db.listMissions()[0]
     if (mission?.githubPrUrl) throw new Error(`Unexpected PR URL recorded: ${mission.githubPrUrl}`)
     if (!db.queryEvents({ entityId: mission!.id, limit: 20 }).some((e) => e.type === 'operator.draft_pr_blocked')) {
@@ -374,6 +388,48 @@ async function assertNoForbiddenVisibleText(page: Page, where: string): Promise<
   }
 }
 
+/**
+ * v6-P2 (GR#11) — assert that the loop card is present with one of the
+ * expected card types, that an ordinary user can read a non-empty next_step
+ * in visible text (no logs needed), and that the card's machine codes stay in
+ * data-* attributes only — never inside its visible text.
+ */
+async function expectLoopCard(page: Page, expectedTypes: string[], where: string, timeoutMs = 20_000): Promise<void> {
+  const card = page.getByTestId('cockpit-loop-card')
+  const deadline = Date.now() + timeoutMs
+  let lastType = 'absent'
+  while (Date.now() < deadline) {
+    if (await card.count()) {
+      lastType = (await card.getAttribute('data-card-type')) ?? 'unknown'
+      if (expectedTypes.includes(lastType)) break
+    }
+    await page.waitForTimeout(200)
+  }
+  if (!expectedTypes.includes(lastType)) {
+    throw new Error(`Expected loop card type in [${expectedTypes.join(', ')}] ${where}, got: ${lastType}`)
+  }
+  const nextStep = oneLine(await page.getByTestId('cockpit-loop-card-next-step').innerText())
+  const nextStepBody = nextStep.replace(/下一步 · Next/g, '').trim()
+  if (nextStepBody.length < 8) {
+    throw new Error(`Loop card next_step is empty/too short ${where}: "${nextStep}"`)
+  }
+  const cardText = await card.innerText()
+  for (const re of FORBIDDEN_VISIBLE) {
+    if (re.test(cardText)) throw new Error(`Raw machine code ${re} leaked into the loop card ${where}`)
+  }
+  for (const attr of ['data-hold-code', 'data-pr-gate-code', 'data-machine-stage'] as const) {
+    const code = await card.getAttribute(attr)
+    // Stage tokens that are plain words (e.g. "running") can legitimately occur
+    // in calm prose; only machine-looking tokens (underscored / HOLD-*) are
+    // forbidden from visible text.
+    const machineLooking = Boolean(code) && (code!.includes('_') || code!.startsWith('HOLD-'))
+    if (code && machineLooking && cardText.includes(code)) {
+      throw new Error(`Machine code "${code}" (${attr}) rendered as visible loop-card text ${where}`)
+    }
+  }
+  note(`loop card ${where}: type=${lastType} · next_step="${nextStepBody.slice(0, 120)}"`)
+}
+
 async function rootStage(page: Page): Promise<string> {
   return (await page.getByTestId('cockpit-root').getAttribute('data-stage')) ?? 'unknown'
 }