feat: 加 prompt-stats 脚本，对比某天前后学生和 AI 对话的情况

轮数分布、层级分布、第 2 轮起的 L1、只粘贴外部代码的比例、“出问题了”“帮我写清楚” 和交作品反思的使用情况。只读。 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
feat: “出问题了”三栏说清问题，交作品时写一句反思，外部提示词至少 8 个字
2026-10-27 22:16:33 +08:00 · 2026-10-27 22:10:48 +08:00 · 2026-10-27 21:40:04 +08:00 · 2026-10-27 21:36:06 +08:00 · 2026-10-27 21:13:46 +08:00 · 2026-10-27 21:09:47 +08:00
23 changed files with 1140 additions and 98 deletions
@@ -17,9 +17,9 @@ The backend was rewritten from Django in 2026-10 to cut memory use (~25 MB vs. 1
  - `src/auth.ts` — Django-compatible password hashing; JWT login, no cookies: `POST /account/login` returns a 15-min HS256 `access_token` (sent as `Authorization: Bearer`, signed with `JWT_SECRET`, random per process if unset) and a 14-day opaque `refresh_token` (only its sha256 is stored, in `auth_refresh_token`; `POST /account/refresh` trades it for a new access token, logout / password change / deactivation delete it). An invalid or expired Bearer token is a 401 even on public endpoints so the client knows to refresh. `requireUser/requireAdmin/requireSuper` (401 / 403)
  - `src/routes/` — `account`, `tasks` (tutorial / challenge / creation), `files` (task assets + image upload), `submission`, `showcase` (mounted under `/submission/showcase`), `review` (mounted under `/submission/review`: teacher class × task wall — no `classname` means all classes —, students' peer works with `scope=class|all`), `prompt`
  - Who can read a submission (`canView` in `routes/submission.ts`): staff, the owner, anyone if it's in the showcase, or any student who has submitted the same task (any class). Students' `GET /submission/` only lists their own. Peer works and the random-rating item are anonymous until the viewer rates them (`PUT /:id/score` returns the author). `flag` now only carries `present` (teacher's in-class presentation queue); the old four colors are no longer shown
-  - `src/services/` — `score.ts` (rating weights super×0.5, admin×0.3, normal×0.2, then Bayesian smoothing with C=3 and zones), `gradebook.ts`, `llm.ts` (OpenAI-compatible streaming via fetch; DeepSeek default, ARK for doubao models; chat context is the last 20 messages with older code blocks elided), `classifier.ts` (Bloom-level L1–L6 prompt tagging, runs in background), `cascade.ts` (Django's CASCADE deletes done by hand, plus the shared delete rules: `hasRatingsOrAwards`, `deleteRound`), `pyround.ts` (`pyRound`/`pySum` reproduce Python's `round()` and `sum()` so scores match the old system bit-for-bit)
+  - `src/services/` — `score.ts` (rating weights super×0.5, admin×0.3, normal×0.2, then Bayesian smoothing with C=3 and zones), `gradebook.ts`, `llm.ts` (OpenAI-compatible streaming via fetch; DeepSeek default, ARK for doubao models; chat context is the last 20 messages with code kept only in the newest reply that has code. Chat replies are written as **做了什么 / 我替你定的 / 想一想** — the decisions the AI made for the student and one open question, to prompt the next round; a reply to a plain question has no code and the WebSocket sends `code: null`. For challenges the AI also gets the task brief (`challengeBrief`) as a checklist only (`BRIEF_RULES`): it builds just what the student asked for and asks about one missing requirement in “想一想”; creations get no brief. The “帮我写清楚” coach gets the task brief, recent prompts and the last AI explanation from `guidanceContext` in `services/prompt.ts`, and judges by what the student added beyond the brief, like the classifier), `classifier.ts` (Bloom-level L1–L6 prompt tagging, runs in background; the model sees the task brief and the previous prompt, temperature 0, prompts under 4 chars get no level), `cascade.ts` (Django's CASCADE deletes done by hand, plus the shared delete rules: `hasRatingsOrAwards`, `deleteRound`), `pyround.ts` (`pyRound`/`pySum` reproduce Python's `round()` and `sum()` so scores match the old system bit-for-bit)
  - `src/ws.ts` — `/ws/prompt/<task_id>/?token=<access_token>` (code generation chat; the AI context and the page's chat include `conversation`, `edit` and `manual` messages) and `/ws/guidance/<task_id>/?token=...` (prompt coaching); one message per connection at a time (a message sent while the AI is still replying gets an error, payloads capped at 64 KB); on disconnect the pending user message is deleted and the LLM request aborted
-  - `scripts/` — one-off CLIs: `create-superuser.ts`, `classify-prompts.ts`, `recalculate-scores.ts`
+  - `scripts/` — one-off CLIs: `create-superuser.ts`, `classify-prompts.ts`, `recalculate-scores.ts`, `prompt-stats.ts` (read-only before/after comparison of AI-chat behavior: rounds, levels, external-code share, “出问题了” / reflection usage)
  - Roles: `super`, `admin`, `normal`. Task model is multi-table: `task_task` + one of `task_tutorial` / `task_challenge` / `task_creation` (`task_ptr_id`)
 - **`client/`** — Vue 3 + TypeScript SPA built with **Vite**; Bun is the package manager and runtime (no Node/npm)
@@ -30,7 +30,7 @@ The backend was rewritten from Django in 2026-10 to cut memory use (~25 MB vs. 1
  - `src/router.ts` — Vue Router; `meta.auth` needs login, `meta.role` (`staff` = admin+super, `super`) is checked in the guard after `store/session.loadUser()`, `meta.bare` pages render without the top bar
  - `src/theme.ts` — the only place colors are defined: one palette (light + dark values) becomes both `--wp-*` CSS variables and Naive UI `themeOverrides`; use `var(--wp-...)` in styles, never hard-coded colors (iframes stay white: they show students' pages). No `color-mix()` (Chrome 105). `store/theme.ts` holds the mode (auto / light / dark, toggled in the top bar) and sets `data-theme` on `<html>`; `styles/dark-content.css` is generated from the GitHub dark markdown and highlight.js styles (selectors prefixed with `:root[data-theme="dark"]`); CodeMirror uses One Dark highlighting in dark mode
  - `src/icons.ts` — `<Icon icon="lucide:name" />` renders bundled `lucide-vue-next` icons (works offline); register new names in its map. Fonts are bundled too (`@fontsource/jetbrains-mono`)
-  - `components/workbench/Workbench.vue` — shared challenge / creation page: left = brief + AI chat (`PromptPanel`), right = preview / code (`CodeTabs`) + `VersionBar`. Every assistant message with code is a version; hand edits and pasted external-AI code are saved as versions via `POST /prompt/versions` (`source` `edit` / `manual`, stored as a user+assistant pair). Challenges are submitted only when the student clicks “交这一版” (`POST /submission/` with `message_id`), never automatically. Editor code is per task (`composables/useTaskCode.ts`, localStorage `web-code:<user>:<task>`); the tutorial page edits the shared `store/editors.ts` refs but `Workspace.vue` saves them under the same per-task key and `TutorialContent` restores them (example code only the first time)
+  - `components/workbench/Workbench.vue` — shared challenge / creation page: left = brief + AI chat (`PromptPanel`), right = preview / code (`CodeTabs`) + `VersionBar`. Every assistant message with code is a version; hand edits and pasted external-AI code are saved as versions via `POST /prompt/versions` (`source` `edit` / `manual`, stored as a user+assistant pair). Challenges are submitted only when the student clicks “交这一版” (`POST /submission/` with `message_id`), never automatically; `SubmitModal` first asks an optional one-line reflection (最满意的改动和理由), stored in `submission_reflection` (`ON DELETE CASCADE`, returned by `GET /submission/:id` only to staff and the author, shown as “学生说” in `ReviewPanel`). In `PromptPanel`, “出问题了” is a three-field form (做了什么 / 想看到什么 / 实际看到了什么) joined into one prompt; a short vague complaint (“还是不行”) gets a one-time hint offering it. Pasted external-AI prompts need at least 8 characters Editor code is per task (`composables/useTaskCode.ts`, localStorage `web-code:<user>:<task>`); the tutorial page edits the shared `store/editors.ts` refs but `Workspace.vue` saves them under the same per-task key and `TutorialContent` restores them (example code only the first time)
  - `pages/ChallengeHome.vue` — challenge list page; `pages/TaskEditor.vue` (prop `kind`) is the single tutorial / challenge authoring page
  - `pages/Review.vue` (“批改”, default teaching page) + `components/review/ReviewPanel.vue` (also opened from the gradebook) — opens as a large preview (wall hidden, Esc returns; “小面板” switches to a side panel, remembered); keys: J/K previous/next student (by key position), 1–5 rate, P present; “全部班级” groups the wall by class; `components/review/PeersDrawer.vue` is the students' “同学作品” (本班 / 全部班级)
  - `components/layout/AppShell.vue` — top bar for every page; `pages/Dashboard.vue` is the teaching sidebar (“教学”) under `/dashboard`
@@ -62,6 +62,7 @@ bun run typecheck
 bun scripts/create-superuser.ts <username> <password>
 bun scripts/classify-prompts.ts [--task-id N] [--force] [--dry-run]
 bun scripts/recalculate-scores.ts [--dry-run]
 bun scripts/prompt-stats.ts <split-date> [--from date] [--to date]   # dates in Beijing time
 ```
 Config comes from `server/.env` (see `server/.env.example`); dev DB is `server/data/db.sqlite3`, uploads go to the repo-root `media/`.
@@ -264,6 +264,7 @@ export const Submission = {
      prompt?: string
    },
    messageId?: number,
    reflection?: string,
  ) {
    const { prompt, ...rest } = code
    const data = {
@@ -271,6 +272,7 @@ export const Submission = {
      ...rest,
      prompt: prompt || null,
      message_id: messageId ?? null,
      reflection: reflection || null,
    }
    const res = await http.post("/submission/", data)
    return res.data
@@ -52,20 +52,59 @@
      <!-- Streaming indicator -->
      <div v-if="streaming" class="message assistant">
        <div class="message-role">AI</div>
-        <div v-if="!streamingContent" class="typing-indicator">
+        <template v-if="!streamingContent">
          <div class="typing-indicator">
            <span></span><span></span><span></span>
          </div>
          <div class="streaming-hint">
            AI 正在构思… 已 {{ elapsedText }}<template v-if="elapsed >= 20">，复杂的网页要想一两分钟</template>
          </div>
        </template>
        <template v-else>
          <div
-          v-else
+            v-if="streamParts.prose"
            class="message-content"
-          v-html="renderMarkdown(streamingContent)"
+            v-html="renderMarkdown(streamParts.prose)"
          ></div>
-        <div class="streaming-hint">AI 正在生成中…</div>
+          <div v-if="streamParts.lines >= 0" class="code-progress">
            <Icon icon="lucide:sparkles" :width="14" />
            正在写网页… 已写 {{ streamParts.lines }} 行
          </div>
          <div v-else class="streaming-hint">AI 正在回复…</div>
        </template>
      </div>
      <RatingCard v-if="rating" :active="streaming" />
    </div>
    <GuidancePanel @generate="onGuidanceGenerate" />
    <div class="input-area">
      <!-- 只说“不行”“没反应”时提醒一下，可以直接发 -->
      <div v-if="vagueHint" class="vague-hint">
        <span>说清楚哪里不对，AI 更容易一次修好</span>
        <n-flex :size="4" :wrap="false">
          <n-button size="tiny" type="primary" secondary @click="openProblem">说清楚</n-button>
          <n-button size="tiny" quaternary @click="sendAnyway">直接发</n-button>
        </n-flex>
      </div>
      <div v-if="problemOpen" class="problem">
        <div class="problem-title">出问题了？说清楚这三件事</div>
        <label>我做了什么</label>
        <n-input v-model:value="problem.did" placeholder="比如：点了「开始」按钮" />
        <label>想看到什么</label>
        <n-input v-model:value="problem.want" placeholder="比如：倒计时开始走" />
        <label>实际看到了什么</label>
        <n-input v-model:value="problem.saw" placeholder="比如：数字一直停在 60 不动" />
        <n-flex justify="end" :size="8" style="margin-top: 10px">
          <n-button @click="problemOpen = false">取消</n-button>
          <n-button
            type="primary"
            :disabled="!problem.did.trim() || !problem.saw.trim() || !connected || streaming"
            @click="sendProblem"
          >
            发给 AI
          </n-button>
        </n-flex>
      </div>
      <template v-else>
        <n-flex
          class="prompt-input-row"
          align="flex-start"
@@ -76,7 +115,7 @@
            v-model:value="input"
            type="textarea"
            :autosize="{ minRows: 1, maxRows: 4 }"
-          placeholder="描述你想要的网页效果..."
+            placeholder="用你自己的话说说想做成什么样…"
            :disabled="streaming"
            style="flex: 1; min-width: 0"
            @keydown.enter.exact.prevent="send"
@@ -98,7 +137,19 @@
            AI 教练会问你几个问题，帮你把想要的效果描述清楚
          </n-tooltip>
        </n-flex>
-      <n-flex justify="flex-end" align="center" style="margin-top: 8px">
+        <n-flex justify="space-between" align="center" style="margin-top: 8px">
          <n-tooltip>
            <template #trigger>
              <n-button quaternary :disabled="!hasVersion || streaming" @click="openProblem">
                <template #icon>
                  <Icon icon="lucide:bug" :width="16" />
                </template>
                出问题了
              </n-button>
            </template>
            网页哪里不对？说清楚做了什么、想看到什么、实际看到了什么
          </n-tooltip>
          <n-flex align="center" :size="8">
            <n-select
              v-model:value="selectedModel"
              :options="modelOptions"
@@ -117,12 +168,14 @@
              发送
            </n-button>
          </n-flex>
        </n-flex>
      </template>
    </div>
  </div>
 </template>
 <script setup lang="ts">
-import { ref, watch, nextTick, computed } from "vue"
+import { ref, reactive, watch, nextTick, computed, onUnmounted } from "vue"
 import { useStorage } from "@vueuse/core"
 import { useMessage } from "naive-ui"
 import { Icon } from "../../icons"
@@ -133,6 +186,7 @@ import {
  messages,
  streaming,
  streamingContent,
  streamStartedAt,
  connected,
  sendPrompt,
  stopPrompt,
@@ -221,6 +275,46 @@ async function deletePair(assistantMsgId: number) {
 async function send() {
  const text = input.value.trim()
  if (!text || streaming.value) return
  // 只说“不行”“没反应”，AI 只能瞎猜；提醒一次，学生选“直接发”以后这次就不再拦
  if (!skipVagueCheck && hasVersion.value && [...text].length <= 15 && VAGUE.test(text)) {
    vagueHint.value = true
    return
  }
  input.value = ""
  await props.beforeSend?.()
  sendPrompt(text, selectedModel.value)
 }
 // “出问题了”：做了什么、想看到什么、实际看到了什么，拼成一句发给 AI
 const VAGUE =
  /不行|不对|没反应|没用|没效果|用不了|不能用|不能玩|动不了|进不去|打不开|没显示|不显示|显示不出|bug|出错|报错|错了|坏了|白屏|空白|有问题/i
 const vagueHint = ref(false)
 let skipVagueCheck = false
 const problemOpen = ref(false)
 const problem = reactive({ did: "", want: "", saw: "" })
 watch(input, () => (vagueHint.value = false))
 const hasVersion = computed(() => messages.value.some((m) => m.role === "assistant" && hasCode(m)))
 function openProblem() {
  vagueHint.value = false
  problemOpen.value = true
 }
 function sendAnyway() {
  skipVagueCheck = true
  vagueHint.value = false
  send()
 }
 async function sendProblem() {
  const part = (s: string) => s.trim().replace(/[。，,.！!；;]+$/, "")
  const did = part(problem.did).replace(/^我/, "")
  const want = part(problem.want)
  const text = `我${did}，${want ? `想看到${want}，` : ""}但实际${part(problem.saw)}。`
  Object.assign(problem, { did: "", want: "", saw: "" })
  problemOpen.value = false
  input.value = ""
  await props.beforeSend?.()
  sendPrompt(text, selectedModel.value)
@@ -238,10 +332,47 @@ async function onGuidanceGenerate(finalPrompt: string) {
  sendPrompt(finalPrompt, selectedModel.value)
 }
-function renderContent(msg: { role: string; content: string }): string {
+// 学生只关心成品：AI 回复里的代码块不在对话里展开（右边“代码”里能看），只说写了多少行。
-  return renderMarkdown(msg.content)
+// 按 ``` 切开，奇数段是代码；段数是偶数说明最后一个代码块还没写完
 function splitCode(text: string) {
  const parts = text.split("```")
  const prose = parts.filter((_, i) => i % 2 === 0).join("\n\n").trim()
  const code = parts.filter((_, i) => i % 2 === 1)
  const lines = code.reduce((n, c) => n + c.split("\n").length - 1, 0)
  return { prose, lines: code.length ? lines : -1 }
 }
 const streamParts = computed(() => splitCode(streamingContent.value))
 function renderContent(msg: { role: string; content: string }): string {
  if (msg.role !== "assistant") return renderMarkdown(msg.content)
  const { prose, lines } = splitCode(msg.content)
  const note = lines >= 0 ? `<p class="code-note">网页代码 ${lines} 行，在右边“代码”里看</p>` : ""
  return renderMarkdown(prose) + note
 }
 // 等 AI 的计时：每秒刷新一次
 const now = ref(Date.now())
 let ticker: ReturnType<typeof setInterval> | undefined
 watch(
  streaming,
  (on) => {
    clearInterval(ticker)
    now.value = Date.now()
    if (on) ticker = setInterval(() => (now.value = Date.now()), 1000)
  },
  { immediate: true },
 )
 onUnmounted(() => clearInterval(ticker))
 const elapsed = computed(() =>
  streamStartedAt.value ? Math.max(0, Math.floor((now.value - streamStartedAt.value) / 1000)) : 0,
 )
 const elapsedText = computed(() => {
  const s = elapsed.value
  return s < 60 ? `${s} 秒` : `${Math.floor(s / 60)} 分 ${s % 60} 秒`
 })
 watch([() => messages.value.length, streamingContent], () => {
  nextTick(() => {
    if (messagesRef.value) {
@@ -377,6 +508,20 @@ watch([() => messages.value.length, streamingContent], () => {
  }
 }
 .code-progress {
  display: flex;
  align-items: center;
  gap: 6px;
  margin-top: 6px;
  font-size: 12px;
  color: var(--wp-muted);
 }
 .message-content :deep(.code-note) {
  font-size: 12px;
  color: var(--wp-muted);
 }
 .streaming-hint {
  font-size: 11px;
  color: var(--wp-faint);
@@ -384,6 +529,31 @@ watch([() => messages.value.length, streamingContent], () => {
  animation: pulse 1.5s ease-in-out infinite;
 }
 .vague-hint {
  display: flex;
  align-items: center;
  justify-content: space-between;
  gap: 8px;
  margin-bottom: 8px;
  padding: 6px 10px;
  border-radius: 6px;
  background: var(--wp-brand-soft);
  font-size: 13px;
 }
 .problem-title {
  margin-bottom: 4px;
  font-size: 13px;
  font-weight: 600;
 }
 .problem label {
  display: block;
  margin: 8px 0 4px;
  font-size: 12px;
  color: var(--wp-muted);
 }
 .input-area {
  flex-shrink: 0;
  padding: 12px;
@@ -119,16 +119,21 @@
          </n-popconfirm>
        </section>
        <section v-if="sub.reflection">
          <h4>学生说</h4>
          <div class="reflection">{{ sub.reflection }}</div>
        </section>
        <section>
          <n-flex justify="space-between" align="center" class="chain-head">
-            <h4>提示词 <span class="muted" v-if="chain.length">· {{ chain.length }} 轮</span></h4>
+            <h4>提示词 <span class="muted" v-if="chainSteps">· {{ chainSteps }} 轮</span></h4>
            <n-button v-if="chain.length" size="tiny" quaternary @click="showChain = true">
              逐轮看网页
            </n-button>
          </n-flex>
          <div v-if="!chain.length" class="muted">这份提交没有关联的提示词</div>
          <ol class="chain">
-            <li v-for="(r, i) in chain" :key="i">
+            <li v-for="(r, i) in chain" :key="i" :class="{ guide: r.source === 'guidance' }">
              <n-flex :size="6" align="center">
                <span class="src">{{ sourceLabel(r.source) }}</span>
                <span v-if="r.prompt_level" class="level">{{ levelLabel(r.prompt_level) }}</span>
@@ -195,6 +200,8 @@ const versions = ref<SubmissionOut[]>([])
 const chain = ref<PromptRound[]>([])
 const rating = ref(false)
 const showChain = ref(false)
 // 引导对话不生成网页，不算轮数
 const chainSteps = computed(() => chain.value.filter((r) => r.source !== "guidance").length)
 let awardsCache: Promise<AwardManageOut[]> | null = null
 const awards = ref<AwardManageOut[]>([])
@@ -477,6 +484,17 @@ h4 {
  flex-wrap: wrap;
 }
 /* 交的时候写的反思 */
 .reflection {
  padding: 6px 10px;
  border-left: 3px solid var(--wp-brand);
  background: var(--wp-brand-soft);
  font-size: 13px;
  line-height: 1.6;
  white-space: pre-wrap;
  word-break: break-word;
 }
 .chain {
  margin: 0;
  padding-left: 20px;
@@ -490,6 +508,12 @@ h4 {
  color: var(--wp-muted);
 }
 /* 不是 list-item，就不占 ol 的序号 */
 .chain .guide {
  display: block;
  color: var(--wp-muted);
 }
 .level {
  font: 600 11px var(--wp-font-mono);
  color: var(--wp-brand);
@@ -9,7 +9,7 @@
    <template #header>
      <div class="head">
        <b>提示词</b>
-        <span v-if="rounds.length" class="muted">{{ rounds.length }} 轮</span>
+        <span v-if="steps" class="muted">{{ steps }} 轮</span>
        <span v-if="title" class="muted title">{{ title }}</span>
      </div>
    </template>
@@ -37,20 +37,37 @@
        style="padding: 80px 0"
      />
      <div v-else class="body">
        <div class="left">
          <div v-if="showTrend" class="trend">
            <span class="muted">层级</span>
            <div class="bars">
              <button
                v-for="b in trend"
                :key="b.i"
                :class="['bar', { on: b.i === selected, none: !b.level }]"
                :style="{ height: `${b.level ? 6 + b.level * 5 : 4}px` }"
                :title="`第 ${nums[b.i]} 轮${b.level ? ' · ' + levelLabel(b.level) : ''}`"
                @click="select(b.i)"
              ></button>
            </div>
          </div>
          <ol class="rounds">
            <li
              v-for="(r, i) in rounds"
              :key="i"
-            :class="['round', { on: i === selected }]"
+              :class="['round', { on: i === selected, guide: isGuide(r) }]"
-            @click="selected = i"
+              @click="isGuide(r) || (selected = i)"
            >
              <div class="round-head">
-              <span class="num">{{ i + 1 }}</span>
+                <span class="num">{{ nums[i] }}</span>
                <span :class="['src', `src-${r.source || 'conversation'}`]">
                  {{ sourceLabel(r.source) }}
                </span>
                <span v-if="r.prompt_level" class="level">{{ levelLabel(r.prompt_level) }}</span>
                <span v-if="repeats[i]" class="repeat" title="和前面某一轮的提问一字不差">重复提问</span>
                <span v-if="changes[i]" class="muted">{{ changes[i] }}</span>
                <span v-if="i === rounds.length - 1" class="final">交的这一版</span>
                <span v-else-if="isGuide(r)" class="muted no-code">问写提示词的建议，不生成网页</span>
                <span v-else-if="!r.html" class="muted no-code">没有网页</span>
              </div>
              <div
@@ -62,10 +79,11 @@
              </button>
            </li>
          </ol>
        </div>
        <div class="stage">
          <div class="stage-head">
-            <b>第 {{ selected + 1 }} 轮生成的网页</b>
+            <b>第 {{ nums[selected] }} 轮生成的网页</b>
            <span class="muted">↑ ↓ 切换</span>
          </div>
          <div class="frame">
@@ -110,6 +128,65 @@ const rounds = ref<PromptRound[]>([])
 const selected = ref(0)
 const expanded = ref(new Set<number>())
 // 引导对话（问怎么写提示词）夹在 AI 对话中间，不算轮数，也不能选中
 const isGuide = (r: PromptRound) => r.source === "guidance"
 const steps = computed(() => rounds.value.filter((r) => !isGuide(r)).length)
 const nums = computed(() => {
  let n = 0
  return rounds.value.map((r) => (isGuide(r) ? "" : String(++n)))
 })
 // 层级走势：每轮 AI 对话 / 外部 AI 一根柱子，两轮以上、有标过层级才显示
 const trend = computed(() =>
  rounds.value.flatMap((r, i) => (isGuide(r) ? [] : [{ i, level: r.prompt_level }])),
 )
 const showTrend = computed(() => trend.value.length > 1 && trend.value.some((b) => b.level))
 // 和前面某一轮一字不差的提问（手改那一轮的正文都一样，不算）
 const repeats = computed(() => {
  const seen = new Set<string>()
  return rounds.value.map((r) => {
    if (isGuide(r) || r.source === "edit") return false
    const key = r.question.replace(/\s+/g, " ").trim()
    const dup = seen.has(key)
    seen.add(key)
    return dup
  })
 })
 // 这一轮的代码比上一版改了多少行：只按行内容数，不管顺序
 function codeLines(r: PromptRound) {
  return [r.html, r.css, r.js]
    .join("\n")
    .split("\n")
    .map((l) => l.trim())
    .filter(Boolean)
 }
 function lineDiff(before: string[], after: string[]) {
  const left = new Map<string, number>()
  for (const l of before) left.set(l, (left.get(l) ?? 0) + 1)
  let added = 0
  for (const l of after) {
    const n = left.get(l) ?? 0
    if (n) left.set(l, n - 1)
    else added++
  }
  const removed = before.length - (after.length - added)
  return added || removed ? `+${added} −${removed} 行` : "代码没变"
 }
 const changes = computed(() => {
  let prev: string[] | null = null
  return rounds.value.map((r) => {
    if (isGuide(r) || !r.html) return ""
    const cur = codeLines(r)
    const text = prev ? lineDiff(prev, cur) : ""
    prev = cur
    return text
  })
 })
 const doc = computed(() => {
  const r = rounds.value[selected.value]
  if (!r?.html) return ""
@@ -154,6 +231,13 @@ async function load() {
  }
 }
 function select(i: number) {
  selected.value = i
  nextTick(() =>
    document.querySelector(".rounds .round.on")?.scrollIntoView({ block: "nearest" }),
  )
 }
 // 在捕获阶段接管按键：弹框开着时，批改页的 Esc / J / K、批改面板的 1–5 / P 都不能触发
 function onKey(e: KeyboardEvent) {
  if (!props.show) return
@@ -165,10 +249,9 @@ function onKey(e: KeyboardEvent) {
  const step = e.key === "ArrowDown" ? 1 : e.key === "ArrowUp" ? -1 : 0
  if (!step || !rounds.value.length) return
  e.preventDefault()
-  selected.value = Math.min(rounds.value.length - 1, Math.max(0, selected.value + step))
+  let i = selected.value + step
-  nextTick(() =>
+  while (rounds.value[i] && isGuide(rounds.value[i]!)) i += step
-    document.querySelector(".rounds .round.on")?.scrollIntoView({ block: "nearest" }),
+  if (rounds.value[i]) select(i)
  )
 }
 watch(
@@ -222,7 +305,57 @@ onUnmounted(() => window.removeEventListener("keydown", onKey, true))
  height: 72vh;
 }
 .left {
  min-height: 0;
  display: flex;
  flex-direction: column;
  gap: 8px;
 }
 .trend {
  display: flex;
  align-items: flex-end;
  gap: 10px;
  padding: 0 6px 6px 2px;
  border-bottom: 1px solid var(--wp-line);
 }
 .bars {
  flex: 1;
  height: 38px;
  display: flex;
  align-items: flex-end;
  gap: 3px;
 }
 .bar {
  flex: 1;
  max-width: 22px;
  padding: 0;
  border: 0;
  border-radius: 2px 2px 0 0;
  background: var(--wp-faint);
  opacity: 0.55;
  cursor: pointer;
 }
 .bar.none {
  background: var(--wp-line);
  opacity: 1;
 }
 .bar:hover,
 .bar.on {
  opacity: 1;
 }
 .bar.on {
  background: var(--wp-brand);
 }
 .rounds {
  flex: 1;
  min-height: 0;
  margin: 0;
  padding: 0 6px 0 0;
  list-style: none;
@@ -269,6 +402,7 @@ onUnmounted(() => window.removeEventListener("keydown", onKey, true))
 .src,
 .level,
 .repeat,
 .final {
  padding: 0 6px;
  border-radius: 4px;
@@ -291,6 +425,30 @@ onUnmounted(() => window.removeEventListener("keydown", onKey, true))
  background: var(--wp-html-soft);
 }
 .src-guidance {
  color: var(--wp-muted);
  background: var(--wp-paper);
 }
 .round.guide {
  border-style: dashed;
  background: transparent;
  cursor: default;
 }
 .round.guide:hover {
  border-color: var(--wp-line);
 }
 .round.guide .q {
  color: var(--wp-muted);
 }
 .repeat {
  color: var(--wp-warn);
  background: var(--wp-warn-soft);
 }
 .level {
  color: var(--wp-ink);
  background: var(--wp-paper);
@@ -130,6 +130,34 @@
            </div>
            <div v-else class="empty">还没有打过分的作品</div>
          </section>
          <!-- 怎么问的 AI -->
          <section class="span">
            <h4>
              怎么问的 AI
              <span class="count">每人最新一版 · {{ stats.prompts.with_chain }} 份有提示词记录</span>
            </h4>
            <div v-if="stats.prompts.with_chain" class="prompt">
              <div class="facts">
                <div><b>{{ stats.prompts.avg_rounds }}</b>平均几轮</div>
                <div><b>{{ pct(stats.prompts.one_round, stats.prompts.with_chain) }}</b>一轮就交</div>
                <div><b>{{ pct(stats.prompts.external, stats.prompts.with_chain) }}</b>最后一版用外部 AI</div>
                <div><b>{{ pct(stats.prompts.used_guidance, stats.prompts.with_chain) }}</b>问过怎么写提示词</div>
              </div>
              <div class="hist">
                <template v-for="(n, i) in stats.prompts.levels" :key="i">
                  <span class="hist-label">{{ levelLabel(i + 1) }}</span>
                  <span class="hist-bar">
                    <span :style="{ width: pct(n, maxLevel) }"></span>
                  </span>
                  <span class="hist-value">{{ n }}</span>
                </template>
              </div>
            </div>
            <div v-else class="empty">
              {{ stats.submitted_count ? "都是直接交的代码，没有提示词记录" : "还没有人交" }}
            </div>
          </section>
        </div>
      </div>
    </n-spin>
@@ -150,6 +178,7 @@ import { computed, ref, watch } from "vue"
 import { useRouter } from "vue-router"
 import { Icon } from "../../icons"
 import { Submission } from "../../api"
 import { levelLabel } from "../../utils/helper"
 import type { TaskStatsOut, TopScoredItem, UserTag } from "../../utils/type"
 // 老师下课前看的这道题的总结：默认看批改页正在看的班
@@ -200,6 +229,7 @@ const buckets = computed(() => {
  ]
 })
 const maxBucket = computed(() => Math.max(...buckets.value.map((b) => b.value)))
 const maxLevel = computed(() => Math.max(0, ...(stats.value?.prompts.levels ?? [])))
 const unsubmittedGroups = computed(() => byClass(stats.value?.unsubmitted_users ?? []))
 const unratedGroups = computed(() => byClass(stats.value?.unrated_users ?? []))
@@ -540,4 +570,29 @@ h4 {
 .go {
  color: var(--wp-faint);
 }
 .span {
  grid-column: 1 / -1;
 }
 .prompt {
  display: grid;
  grid-template-columns: 1fr 1fr;
  gap: 20px;
 }
 .facts {
  display: grid;
  grid-template-columns: 1fr 1fr;
  gap: 10px 16px;
  align-content: start;
  font-size: 12px;
  color: var(--wp-muted);
 }
 .facts b {
  display: block;
  font: 600 18px/1.3 var(--wp-font-mono);
  color: var(--wp-ink);
 }
 </style>
@@ -5,12 +5,16 @@
    title="粘贴外部 AI 生成的代码"
    style="width: 640px; max-width: 94vw"
  >
-    <n-form-item label="你在外部 AI 里用的提示词" :show-feedback="false">
+    <n-form-item
      label="你在外部 AI 里说了什么"
      :validation-status="promptTooShort ? 'warning' : undefined"
      :feedback="promptTooShort ? `再写具体一点：你让 AI 做了什么、改了什么（至少 ${MIN_PROMPT} 个字）` : ''"
    >
      <n-input
        v-model:value="prompt"
        type="textarea"
        :autosize="{ minRows: 2, maxRows: 6 }"
-        placeholder="把你发给豆包、DeepSeek 等 AI 的话粘贴到这里"
+        placeholder="把你发给豆包、DeepSeek 等 AI 的话粘贴到这里；说了好几轮的，每一轮都贴进来，一轮一行"
      />
    </n-form-item>
    <n-form-item label="完整的网页代码" :show-feedback="false" style="margin-top: 12px">
@@ -30,7 +34,7 @@
        <n-button @click="show = false">取消</n-button>
        <n-button
          type="primary"
-          :disabled="!prompt.trim() || !raw.trim()"
+          :disabled="[...prompt.trim()].length < MIN_PROMPT || !raw.trim()"
          :loading="saving"
          @click="save"
        >
@@ -41,7 +45,7 @@
  </n-modal>
 </template>
 <script lang="ts" setup>
-import { ref } from "vue"
+import { computed, ref } from "vue"
 import { splitHtml, type Code } from "../../utils/code"
 const show = defineModel<boolean>("show", { required: true })
@@ -50,6 +54,12 @@ const props = defineProps<{
 }>()
 const prompt = ref("")
 // 以前四分之一的外部提示词只有“1”“倒计时”这样几个字，看不出学生想了什么
 const MIN_PROMPT = 8
 const promptTooShort = computed(() => {
  const n = [...prompt.value.trim()].length
  return n > 0 && n < MIN_PROMPT
 })
 const raw = ref("")
 const saving = ref(false)
@@ -0,0 +1,67 @@
 <template>
  <n-modal
    v-model:show="show"
    preset="card"
    :title="title"
    style="width: 520px; max-width: 94vw"
    :mask-closable="!saving"
  >
    <div class="question">
      {{ hasEarlier ? "比起第一版，这一版你最满意的改动是什么？为什么这样改？" : "这一版你最满意的地方是什么？为什么？" }}
    </div>
    <n-input
      v-model:value="reflection"
      type="textarea"
      :autosize="{ minRows: 3, maxRows: 6 }"
      :maxlength="500"
      placeholder="比如：我把按钮从五个减到三个，因为按钮太多了看不出先点哪个"
    />
    <n-text depth="3" class="hint">可以不写，直接交。写了的话，老师批改时能看到。</n-text>
    <template #footer>
      <n-flex justify="end">
        <n-button :disabled="saving" @click="show = false">取消</n-button>
        <n-button type="primary" :loading="saving" @click="save">交</n-button>
      </n-flex>
    </template>
  </n-modal>
 </template>
 <script lang="ts" setup>
 import { ref } from "vue"
 // 交作品前问一句反思：最满意的改动和理由。不拦着交，只是让学生回头看一眼自己改了什么
 const show = defineModel<boolean>("show", { required: true })
 const props = defineProps<{
  title: string
  // 之前还有别的版本，就问“比起第一版”
  hasEarlier: boolean
  onSubmit: (reflection: string) => Promise<void>
 }>()
 const reflection = ref("")
 const saving = ref(false)
 async function save() {
  saving.value = true
  try {
    await props.onSubmit(reflection.value.trim())
    reflection.value = ""
    show.value = false
  } finally {
    saving.value = false
  }
 }
 </script>
 <style scoped>
 .question {
  margin-bottom: 10px;
  font-size: 14px;
  line-height: 1.6;
  color: var(--wp-ink);
 }
 .hint {
  display: block;
  margin-top: 8px;
  font-size: 12px;
 }
 </style>
@@ -103,6 +103,12 @@
    </template>
  </n-split>
  <PasteCodeModal v-model:show="showPaste" :on-save="savePasted" />
  <SubmitModal
    v-model:show="showSubmit"
    :title="submitTitle"
    :has-earlier="submitHasEarlier"
    :on-submit="submitChallenge"
  />
  <n-drawer v-model:show="showHistory" :width="420" placement="right">
    <n-drawer-content title="全部历史" :native-scrollbar="false" body-content-style="padding: 0">
      <PromptHistoryPanel
@@ -128,6 +134,7 @@ import Preview from "../editor/Preview.vue"
 import CodeTabs from "./CodeTabs.vue"
 import VersionBar, { type VersionItem } from "./VersionBar.vue"
 import PasteCodeModal from "./PasteCodeModal.vue"
 import SubmitModal from "./SubmitModal.vue"
 import { Prompt, Submission } from "../../api"
 import { useTaskCode } from "../../composables/useTaskCode"
 import {
@@ -174,6 +181,7 @@ const selectedId = ref<number | null>(null)
 const busy = ref(false)
 const briefOpen = ref(true)
 const showPaste = ref(false)
 const showSubmit = ref(false)
 const showHistory = ref(false)
 const historyKey = ref(0)
@@ -308,21 +316,46 @@ function openHistory() {
 }
 async function submit() {
-  busy.value = true
+  if (props.kind === "challenge") {
-  try {
+    showSubmit.value = true
    if (props.kind === "creation") {
      await props.saveWork?.({ ...code })
      message.success("作品已保存")
    return
  }
  busy.value = true
  try {
    await props.saveWork?.({ ...code })
    message.success("作品已保存")
  } catch (err: any) {
    message.error(err.response?.data?.detail ?? "保存失败，请重试")
  } finally {
    busy.value = false
  }
 }
 const submitTitle = computed(() => {
  if (modified.value || !selectedId.value) return "交当前修改"
  const n = versions.value.find((v) => v.id === selectedId.value)?.n
  return n ? `交 v${n}` : "交这一版"
 })
 // 要交的这一版之前还有版本（当前修改会存成新的一版）
 const submitHasEarlier = computed(() => {
  if (modified.value || !selectedId.value) return versions.value.length > 0
  return (versions.value.find((v) => v.id === selectedId.value)?.n ?? 1) > 1
 })
 // 挑战：先在弹框里写一句反思（可以不写）再交；出错时抛出去，弹框不关
 async function submitChallenge(reflection: string) {
  busy.value = true
  try {
    const id =
      modified.value || !selectedId.value ? await saveAsVersion() : selectedId.value
-    await Submission.create(props.taskId, { ...code }, id)
+    await Submission.create(props.taskId, { ...code }, id, reflection)
    markMessageSubmitted(id)
    const n = versions.value.find((v) => v.id === id)?.n
    message.success(n ? `已提交 v${n}` : "已提交")
  } catch (err: any) {
    message.error(err.response?.data?.detail ?? "提交失败，请重试")
    throw err
  } finally {
    busy.value = false
  }
@@ -4,6 +4,7 @@ import {
  ArrowRight,
  Award,
  BookOpen,
  Bug,
  ChartNoAxesColumn,
  Check,
  ChevronDown,
@@ -49,6 +50,7 @@ const registry: Record<string, Component> = {
  "arrow-right": ArrowRight,
  award: Award,
  "book-open": BookOpen,
  bug: Bug,
  "bar-chart-2": ChartNoAxesColumn,
  check: Check,
  "chevron-down": ChevronDown,
@@ -19,6 +19,8 @@ export const conversationId = ref<string>("")
 export const connected = ref(false)
 export const streaming = ref(false)
 export const streamingContent = ref("")
 // 这次提问发出的时间：AI 先思考，第一个字可能一两分钟后才出来，界面据此显示等了多久
 export const streamStartedAt = ref(0)
 export const currentTaskId = ref(0)
 // 新代码交给工作台处理：init 是连上时对话里最新的那一版，complete 是 AI 刚生成的
 let _codeSink:
@@ -106,6 +108,7 @@ export function disconnectPrompt() {
 export function sendPrompt(content: string, model: string = "") {
  if (!socket.send({ type: "message", content, model })) return
  streaming.value = true
  streamStartedAt.value = Date.now()
  messages.value.push({ role: "user", content })
 }
@@ -21,17 +21,18 @@ export function goHome(router: any, type: TASK_TYPE, display: number) {
 export function sourceLabel(source: string | null | undefined) {
  if (source === "edit") return "手改"
  if (source === "manual") return "外部 AI"
  if (source === "guidance") return "引导"
  return "AI 对话"
 }
 // 提问的 Bloom 等级（services/classifier.ts 后台打的标签）
 export const PROMPT_LEVELS: Record<number, { name: string; desc: string }> = {
-  1: { name: "记忆", desc: "复述或识别知识点" },
+  1: { name: "记忆", desc: "只说要做什么或只报结果，没有自己的要求和解释" },
-  2: { name: "理解", desc: "用自己的话解释概念" },
+  2: { name: "理解", desc: "想弄懂原理：问为什么、是什么意思、怎么做到的" },
-  3: { name: "应用", desc: "把知识用到新情境" },
+  3: { name: "应用", desc: "提出具体要求（颜色、布局、功能），照搬题目要求也算" },
-  4: { name: "分析", desc: "拆解结构、找出规律" },
+  4: { name: "分析", desc: "说清在什么操作、哪个位置出了什么问题，或推测原因" },
-  5: { name: "评价", desc: "按标准作出判断" },
+  5: { name: "评价", desc: "带理由评判现有结果，或比较不同做法再选择" },
-  6: { name: "创造", desc: "整合信息做出新东西" },
+  6: { name: "创造", desc: "题目之外自己构思的整体设计：主题、玩法、交互" },
 }
 export function levelLabel(level: number | null | undefined) {
@@ -167,6 +167,8 @@ export interface SubmissionAll {
  view_count: number
  created: Date
  modified: Date
  // 交的时候写的反思，只有老师和作者本人拿得到
  reflection?: string | null
 }
 export interface UserTag {
@@ -199,6 +201,18 @@ export interface TaskStatsOut {
  present_count: number
  classes: string[]
  top_scored: TopScoredItem[]
  prompts: PromptChainStats
 }
 // 课堂总结里每人最新一版的提示词链概况
 export interface PromptChainStats {
  with_chain: number
  one_round: number
  external: number
  used_guidance: number
  avg_rounds: number
  // L1–L6 各有几次提问
  levels: number[]
 }
 export type GradebookTaskType = "tutorial" | "challenge"
@@ -0,0 +1,109 @@
 /**
 * 对比某个日期前后，学生和 AI 对话的情况（比如改了 AI 的回复方式以后有没有变化）。
 * 日期按北京时间，分界日期那天算“之后”。只读，不改数据。
 *
 *   bun scripts/prompt-stats.ts <分界日期> [--from 日期] [--to 日期]
 *   bun scripts/prompt-stats.ts 2026-10-28 --from 2026-09-01
 *
 * 层级是打标签时的标准；分类标准改过以后，先跑 classify-prompts.ts --force 重打，前后才好比。
 */
 import { all, toDbTime } from "../src/db"
 const args = process.argv.slice(2)
 const opt = (name: string) => {
  const i = args.indexOf(name)
  return i >= 0 ? args[i + 1] : undefined
 }
 const split = args[0]
 if (!split || !/^\d{4}-\d{2}-\d{2}$/.test(split)) {
  console.log("用法：bun scripts/prompt-stats.ts <分界日期> [--from 日期] [--to 日期]")
  process.exit(1)
 }
 const day = (d: string) => toDbTime(new Date(`${d}T00:00:00+08:00`))
 const nextDay = (d: string) => toDbTime(new Date(new Date(`${d}T00:00:00+08:00`).getTime() + 86_400_000))
 const periods = [
  { name: `${opt("--from") ?? "最早"} ~ ${split} 之前`, from: opt("--from") ? day(opt("--from")!) : "", to: day(split) },
  { name: `${split} ~ ${opt("--to") ?? "现在"}`, from: day(split), to: opt("--to") ? nextDay(opt("--to")!) : "9999" },
 ]
 // 和 PromptPanel 里一样：只说“不行”“没反应”的短抱怨；“出问题了”拼出来的句子
 const VAGUE =
  /不行|不对|没反应|没用|没效果|用不了|不能用|不能玩|动不了|进不去|打不开|没显示|不显示|显示不出|bug|出错|报错|错了|坏了|白屏|空白|有问题/i
 const PROBLEM = /^我.+，(想看到.+，)?但实际.+。$/s
 type Msg = {
  conversation_id: string
  source: string
  content: string
  prompt_level: number | null
  created: string
  // 在这个对话的 AI 对话里是第几轮（只数 conversation）
  round: number | null
 }
 // 批处理脚本，整表扫一遍没关系
 const msgs = all<Msg>(
  `SELECT conversation_id, source, content, prompt_level, created,
     CASE WHEN source = 'conversation'
       THEN ROW_NUMBER() OVER (PARTITION BY conversation_id, source = 'conversation' ORDER BY created, id) END AS round
   FROM prompt_message WHERE role = 'user'`,
 )
 const subs = all<{ created: string; reflected: number }>(
  `SELECT s.created, r.submission_id IS NOT NULL AS reflected
   FROM submission_submission s JOIN task_task t ON t.id = s.task_id
   LEFT JOIN submission_reflection r ON r.submission_id = s.id
   WHERE t.task_type = 'challenge'`,
 )
 const pct = (n: number, d: number) => (d ? `${((n / d) * 100).toFixed(1)}%` : "—")
 const chars = (s: string) => [...s.trim()].length
 function stats(from: string, to: string) {
  const inRange = (t: string) => t >= from && t < to
  const ms = msgs.filter((m) => inRange(m.created))
  const byConv = new Map<string, Msg[]>()
  for (const m of ms) byConv.set(m.conversation_id, [...(byConv.get(m.conversation_id) ?? []), m])
  const convs = [...byConv.values()]
  const ai = convs.filter((c) => c.some((m) => m.source === "conversation"))
  const manualOnly = convs.filter((c) => c.some((m) => m.source === "manual") && !c.some((m) => m.source === "conversation"))
  const rounds = ai.map((c) => c.filter((m) => m.source === "conversation").length)
  const bucket = (lo: number, hi: number) => rounds.filter((n) => n >= lo && n <= hi).length
  const asked = ms.filter((m) => m.source === "conversation")
  const levels = [1, 2, 3, 4, 5, 6].map((l) => asked.filter((m) => m.prompt_level === l).length)
  const leveled = levels.reduce((a, b) => a + b, 0)
  const later = asked.filter((m) => (m.round ?? 1) >= 2 && m.prompt_level)
  const vague = asked.filter((m) => chars(m.content) <= 15 && VAGUE.test(m.content))
  const problem = asked.filter((m) => PROBLEM.test(m.content.trim()))
  const manual = ms.filter((m) => m.source === "manual")
  const s = subs.filter((x) => inRange(x.created))
  return [
    ["有对话或版本的作品", `${convs.length}`],
    ["只粘贴外部代码", `${manualOnly.length}（${pct(manualOnly.length, convs.length)}）`],
    ["和平台 AI 对话的作品", `${ai.length}`],
    ["  平均轮数", ai.length ? (rounds.reduce((a, b) => a + b, 0) / ai.length).toFixed(2) : "—"],
    ["  1 轮", pct(bucket(1, 1), ai.length)],
    ["  2–3 轮", pct(bucket(2, 3), ai.length)],
    ["  4–6 轮", pct(bucket(4, 6), ai.length)],
    ["  7 轮以上", pct(bucket(7, Infinity), ai.length)],
    ["提问（AI 对话）", `${asked.length}，已分级 ${leveled}`],
    ...levels.map((n, i) => [`  L${i + 1}`, pct(n, leveled)]),
    ["  L4–L6", pct(levels[3]! + levels[4]! + levels[5]!, leveled)],
    ["  第 2 轮起的 L1", pct(later.filter((m) => m.prompt_level === 1).length, later.length)],
    ["  只说“不行”的短抱怨", `${vague.length}（${pct(vague.length, asked.length)}）`],
    ["  用“出问题了”", `${problem.length}（${pct(problem.length, asked.length)}）`],
    ["用了“帮我写清楚”的作品", `${convs.filter((c) => c.some((m) => m.source === "guidance")).length}（占和 AI 对话的 ${pct(ai.filter((c) => c.some((m) => m.source === "guidance")).length, ai.length)}）`],
    ["外部 AI 提示词", `${manual.length}，12 字以内 ${pct(manual.filter((m) => chars(m.content) <= 12).length, manual.length)}`],
    ["挑战提交", `${s.length}，写了反思 ${pct(s.filter((x) => x.reflected).length, s.length)}`],
  ]
 }
 const [a, b] = periods.map((p) => stats(p.from, p.to))
 // 终端里中文占两格
 const cols = (k: string) => [...k].reduce((n, ch) => n + (ch.codePointAt(0)! >= 0x2e80 ? 2 : 1), 0)
 const width = Math.max(...a!.map(([k]) => cols(k!))) + 2
 const pad = (k: string) => k + " ".repeat(width - cols(k))
 console.log(`${pad("")}${periods[0]!.name}  →  ${periods[1]!.name}`)
 for (const [i, [k, v]] of a!.entries()) console.log(`${pad(k!)}${v}  →  ${b![i]![1]}`)
@@ -47,6 +47,12 @@ CREATE INDEX IF NOT EXISTS "prompt_conversation_user_id_ecc418b3" ON "prompt_con
 CREATE TABLE IF NOT EXISTS "prompt_message" ("id" integer NOT NULL PRIMARY KEY AUTOINCREMENT, "role" varchar(10) NOT NULL, "content" text NOT NULL, "code_html" text NULL, "code_css" text NULL, "code_js" text NULL, "created" datetime NOT NULL, "conversation_id" char(32) NOT NULL REFERENCES "prompt_conversation" ("id") DEFERRABLE INITIALLY DEFERRED, "prompt_level" integer NULL, "source" varchar(12) NOT NULL, "submission_id" char(32) NULL UNIQUE REFERENCES "submission_submission" ("id") DEFERRABLE INITIALLY DEFERRED);
 CREATE INDEX IF NOT EXISTS "prompt_message_conversation_id_af18dd4b" ON "prompt_message" ("conversation_id");
 CREATE INDEX IF NOT EXISTS "prompt_message_prompt_level_e461b596" ON "prompt_message" ("prompt_level");
 -- 课堂总结统计提示词链用的覆盖索引：只读小列，不碰 content / code_*
 CREATE INDEX IF NOT EXISTS "prompt_message_chain" ON "prompt_message" ("conversation_id", "created", "id", "role", "source", "prompt_level");
 CREATE INDEX IF NOT EXISTS "prompt_message_submission" ON "prompt_message" ("submission_id", "conversation_id", "created", "source") WHERE "submission_id" IS NOT NULL;
 -- 学生交作品时写的一句反思（最满意的改动和理由）；可以不写。删提交时跟着删
 CREATE TABLE IF NOT EXISTS "submission_reflection" ("submission_id" char(32) NOT NULL PRIMARY KEY REFERENCES "submission_submission" ("id") ON DELETE CASCADE, "content" text NOT NULL, "modified" datetime NOT NULL);
 -- 登录用 JWT：access token 不落库，refresh token 只存 sha256
 DROP TABLE IF EXISTS "auth_session";
@@ -26,6 +26,7 @@ import {
  getActiveConversation,
  getPrecedingUserMessage,
  insertMessage,
  promptChainStats,
  type MessageRow,
 } from "../services/prompt"
 import { pyRound } from "../services/pyround"
@@ -119,6 +120,7 @@ router.post("/", async (c) => {
      js: z.string().nullable().optional(),
      prompt: z.string().nullable().optional(),
      message_id: z.number().int().nullable().optional(), // 关联的 assistant message
      reflection: z.string().max(1000).nullable().optional(), // 交的时候写的一句反思，可以不写
    }),
  )
  const task = orNotFound(
@@ -130,6 +132,8 @@ router.post("/", async (c) => {
  const code = { html: p.html ?? null, css: p.css ?? null, js: p.js ?? null }
  const reflection = p.reflection?.trim() || null
  const { submissionId, newUserMsgId } = tx(() => {
    let manualAsstMsgId: number | null = null
    let linkedMsg: MessageRow | null = null
@@ -202,6 +206,15 @@ router.post("/", async (c) => {
    return { submissionId: id, newUserMsgId }
  })
  // 同一版再交时没写就留着上次写的
  if (reflection) {
    run(
      `INSERT INTO submission_reflection (submission_id, content, modified) VALUES ($id, $content, $ts)
       ON CONFLICT (submission_id) DO UPDATE SET content = excluded.content, modified = excluded.modified`,
      { id: submissionId, content: reflection, ts: now() },
    )
  }
  // 只对新增的提示词做层级分类
  if (
    newUserMsgId !== null &&
@@ -473,6 +486,8 @@ router.get("/stats/:task_id", (c) => {
    present_count: studentSubs.filter((s) => s.flag === "present").length,
    classes,
    top_scored: topScored,
    // 每人最新一版是怎么问出来的
    prompts: promptChainStats([...latest.values()].map((s) => s.id)),
  })
 })
@@ -537,7 +552,13 @@ router.get("/:id", (c) => {
  const id = uuidParam(c, "id")
  requireView(user, id)
  const s = orNotFound(get(`${SUBMISSION_SELECT} ${SUBMISSION_FROM} WHERE s.id = $id`, { me: user.id, id }))
-  const out = submissionOut(s)
+  // 反思只给老师和作者本人看
  const reflection =
    isStaff(user) || s.user_id === user.id
      ? (get<{ content: string }>(`SELECT content FROM submission_reflection WHERE submission_id = $id`, { id })
          ?.content ?? null)
      : null
  const out = { ...submissionOut(s), reflection }
  // 和同学作品一样：学生给同学打分之前不告诉作者和得分（作品展里的本来就署名）
  if (!isStaff(user) && s.user_id !== user.id && !out.my_score && !inShowcase(id)) {
    return c.json({ ...out, userid: null, username: null, score: null })
@@ -1,18 +1,32 @@
 import { all, get, run } from "../db"
 import { complete, llmConfigured } from "./llm"
 import { briefOf } from "./prompt"
-const CLASSIFY_SYSTEM_PROMPT = `你是一个教育评估专家。根据布鲁姆认知分类学，分析以下学生在前端学习中发送给AI助手的一条提示词，判断该提示词所体现的认知层级。
+// 学生在这里是“指挥 AI 做网页”，看的是这句提问里学生自己想了多少，不是要 AI 做的东西有多大：
 // 把题目要求原样贴过去，功能列得再多也只是照着做（L3）
 const CLASSIFY_SYSTEM_PROMPT = `你是一个教育评估专家。学生在网页制作课上让 AI 生成或修改网页，下面给出题目要求、学生上一轮的提问和这一轮的提问。请按布鲁姆认知分类学，判断“这一轮提问”体现的学生自己的思考层级。
-层级定义：
+按下面的顺序判断，命中哪一条就是哪一级：
- L1 记忆：能背诵HTML标签语法（例："帮我写一个按钮"）
+1. L1 记忆：没有任何具体要求、解释或问题。只给作品名称、只报结果或只让 AI 修（例："帮我做一个象棋游戏"、"制作一个吃豆人"、"继续"、"预览代码"、"还是没有显示"、"点击开始游戏进不去"、"帮我修复代码bug"）
- L2 理解：能解释flex布局原理（例："为什么这里不居中？"）
+2. L2 理解：想弄懂原理、代码或概念（问为什么、是什么意思、怎么做到的），或者用自己的话解释现有代码在做什么（例："为什么点了按钮没反应？"、"grid 和 flex 有什么区别"、"这段 js 里的 setInterval 是干什么的"）
- L3 应用：能独立搭建页面结构（例："用flex做导航栏，间距16px"）
+3. L4 分析：说清在哪个操作、哪个位置出了什么具体问题，或者自己推测原因、指出问题出在哪部分（例："拖动绘制时松开鼠标还在画，是不是没监听 mouseup"、"缩小窗口后网格就不显示了"、"围棋我刚下一颗黑棋就判我赢了"）
- L4 分析：能定位跨浏览器兼容性bug（例："Safari中margin失效，原因？"）
+4. L5 评价：明确给出判断理由来评价现有结果，或者比较两种做法再选择；必须有“因为……所以……”“A 比 B 好”这样的理由或比较（例："现在按钮太多，小学生看不懂，改成只留三个主要按钮"、"用 canvas 画比用一堆 div 流畅，改成 canvas"）
- L5 评价：能对比并选择方案（例："对比Grid与Flex方案优劣"）
+5. L6 创造：在题目要求之外提出自己构思的设计，而且几部分互相配合成一个整体，如原创的主题、玩法、规则或交互（例："改成星空主题：转盘是星座，抽中稀有奖励时流星划过并解锁对应星座的图鉴"）
- L6 创造：能设计并实现原创交互作品（例："设计夜间/日间切换效果"）
+6. L3 应用：其余给出具体、可执行要求的，如颜色、风格、布局、要加或要改的功能（例："方块大一点，旁边加一个橡皮擦和清空按钮"、"页面风格可爱，浅粉加浅蓝的渐变"、"加上撤销和保存为图片"）
 注意：
 - 只看学生自己想出来的部分。照搬或改写题目要求里的功能清单是 L3，要求写得长、功能多不等于层级高。
 - 只说想要什么风格、加什么功能，没有理由，是 L3 不是 L5；只列很多功能，没有自己的整体构思，是 L3 不是 L6。
 - 只说“没反应”“进不去”“不能玩”“有 bug”，哪怕提到了是哪个按钮，也是 L1；要说出更具体的现象（如“打字速度始终是 0”）或推测原因才是 L4。
 - 说明要改成什么样（如“弧线应该对着中圈”“球门要比禁区小”）是 L3，不是 L4。
 - 整段贴了题目要求或一长串功能清单，不是 L1，是 L3。
 - 结合上一轮看：重复上一轮的提问，层级不高于上一轮。
 只返回一个数字（1-6），不要解释。`
 const MIN_PROMPT_CHARS = 4
 const PROMPT_CHARS = 1500
 const sleep = (ms: number) => new Promise((r) => setTimeout(r, ms))
 async function callLlm(content: string): Promise<number | null> {
@@ -23,7 +37,8 @@ async function callLlm(content: string): Promise<number | null> {
        { role: "system", content: CLASSIFY_SYSTEM_PROMPT },
        { role: "user", content },
      ],
-      { maxTokens: 10, timeoutMs: 30_000 },
+      // 只回一个数字，不用思考；关了思考 temperature 才生效，同一句提示词得到同一个层级
      { thinking: false, maxTokens: 10, temperature: 0, timeoutMs: 30_000 },
    )
    const match = /[1-6]/.exec(text)
    if (!match) {
@@ -37,13 +52,45 @@ async function callLlm(content: string): Promise<number | null> {
  }
 }
 function clip(text: string) {
  return text.length > PROMPT_CHARS ? text.slice(0, PROMPT_CHARS) + "……" : text
 }
 // 发给模型的：题目要求、同一对话里上一轮的提问（引导对话只和引导对话比）、这一轮的提问
 function classifyInput(msg: { id: number; content: string; source: string; conversation_id: string; created: string }) {
  const task = get<{ title: string; content: string }>(
    `SELECT t.title, t.content FROM prompt_conversation c JOIN task_task t ON t.id = c.task_id WHERE c.id = $conv`,
    { conv: msg.conversation_id },
  )
  const prev = get<{ content: string }>(
    `SELECT content FROM prompt_message
     WHERE conversation_id = $conv AND role = 'user' AND source != 'edit'
       AND (source = 'guidance') = $guidance
       AND (created < $created OR (created = $created AND id < $id))
     ORDER BY created DESC, id DESC LIMIT 1`,
    { conv: msg.conversation_id, guidance: msg.source === "guidance", created: msg.created, id: msg.id },
  )
  return [
    `题目：${task?.title ?? "（未知）"}`,
    `题目要求：${task ? briefOf(task.content) : "（无）"}`,
    `上一轮提问：${prev ? clip(prev.content) : "（这是第一轮）"}`,
    `这一轮提问：${clip(msg.content)}`,
  ].join("\n\n")
 }
 export async function classifyMessage(messageId: number): Promise<number | null> {
-  const msg = get<{ content: string }>(
+  const msg = get<{ id: number; content: string; source: string; conversation_id: string; created: string }>(
-    `SELECT content FROM prompt_message WHERE id = $id AND role = 'user' AND source != 'edit'`,
+    `SELECT id, content, source, conversation_id, created FROM prompt_message
     WHERE id = $id AND role = 'user' AND source != 'edit'`,
    { id: messageId },
  )
  if (!msg) return null
-  const level = await callLlm(msg.content)
+  // “1”“，”“可以”这类太短，判断不出层级；清掉以前打的标签（--force 重标时）
  if ([...msg.content.trim()].length < MIN_PROMPT_CHARS) {
    run(`UPDATE prompt_message SET prompt_level = NULL WHERE id = $id`, { id: messageId })
    return null
  }
  const level = await callLlm(classifyInput(msg))
  if (level !== null) {
    run(`UPDATE prompt_message SET prompt_level = $level WHERE id = $id`, {
      level,
@@ -1,18 +1,29 @@
 import { config } from "../config"
-export const SYSTEM_PROMPT = `你是一个网页生成助手。根据用户的需求描述，生成网页代码。
+// AI 回复里的说明是学生唯一会读的文字，用它引导学生想下一步：
 // “我替你定的”指出学生没说、AI 自己拿的主意，学生看了才会说“不要这样，因为……”；“想一想”只问不答
 export const SYSTEM_PROMPT = `你是一个网页生成助手，和学生一起做网页。根据学生的描述生成网页代码。
 规则：
-1. 使用一个 \`\`\`html 代码块返回所有代码
+1. 生成或修改网页时，使用一个 \`\`\`html 代码块返回所有代码
 2. HTML 代码只需要 body 内的内容，不需要完整的 HTML 文档结构
 3. CSS 样式写在 <style> 标签内，JavaScript 写在 <script> 标签内，都放在代码块里
-4. 用中文回复，先简要说明你做了什么，然后给出代码
+4. 在已有代码基础上修改时，返回完整的修改后代码，不要只返回片段
-5. 在已有代码基础上修改时，返回完整的修改后代码，不要只返回片段
+5. 由于任何外部链接都被屏蔽，使用纯 HTML、CSS 和 JS 实现功能，不要依赖外部库
-6. 由于任何外部链接都被屏蔽，使用纯 HTML、CSS 和 JS 实现功能，不要依赖外部库
+6. 用中文回复，对学生称“你”。先写说明，再给代码。说明分下面三项，标题照抄，每项一两句话，不要写长：
   - **做了什么**：这次按学生的要求做了哪些改动
   - **我替你定的**：学生没说、你自己拿主意的地方，比如配色、布局、规则、数值、文字内容；列 1-3 条，只写在网页上看得到、学生可以要求改的。学生这次已经说清楚、没有需要你拿主意的地方，就省略这一项
   - **想一想**：问学生一个问题，帮他想清楚下一步想要什么。问他自己的想法或判断（如“……的时候你希望它怎样？”“这两种你觉得哪个更适合……，为什么？”），不要列一串功能让他挑，也不要替他回答
 7. 学生只是在问问题（为什么、是什么意思、怎么做到的），没有要求改网页时，用浅显的话直接回答，不给代码，也不用分三项
 8. 学生只说“不行”“有 bug”“没反应”，没说清哪里不对时，照样尽力修好，并在“想一想”里请他下次说清楚：做了什么操作、想看到什么、实际看到了什么
 输出格式示例（必须严格遵守，只用一个代码块）：
-好的，我为你创建了一个点击按钮变色的示例。
+**做了什么**：做了一个按钮，点一下会变成红色。
 **我替你定的**：按钮放在页面左上角；变色后不会再变回来。
 **想一想**：再点一次的时候，你希望它变回原来的颜色，还是换成别的颜色？
 \`\`\`html
 <style>
@@ -28,23 +39,36 @@ document.getElementById('btn').onclick = function() {
 </script>
 \`\`\``
-export const GUIDANCE_SYSTEM_PROMPT = `你是一个提示词写作教练，帮助学生写出清晰、具体的网页需求描述。
+// 挑战题目要求只当对照清单：学生没说的不替他做，挑一项放进“想一想”问他打算怎么做，
 // 要做成什么样还是学生自己说；整段贴了题目的照做（不拦，拦了只会去用外部 AI）
 export const BRIEF_RULES = `下面是这道题的题目要求。它是对照清单，不是让你照着做——这是课堂练习，题目要求要学生自己一步步提出来，这正是这节课要练的：
 - 只做学生在对话里说了的内容，做出来要能打开、能用。学生说的话里包含了题目要求（比如整段贴了题目），就照着做
 - 题目要求里学生还没提到的功能、规则和细节（如计时方式、对错的颜色、计分、重来按钮），哪怕很常见、顺手就能做，也先不做；也不要照搬题目里的类名、变量名和数据
 - 从学生没提到的要求里挑一项最重要的放进“想一想”，问他打算怎么做（如“题目还要求标出打错的字，你希望打错的时候是什么样子？”）
 - 学生要的和题目要求冲突时，按学生说的做，在“想一想”里提醒他题目是怎么要求的
 - 题目要求都做到了，“想一想”就问他自己的想法，不再提题目`
-判断标准——满足以下条件视为"够好"：
+// 判断标准和分类器一致：看学生自己想了什么，照搬题目写得再长也不算清楚
- 有明确主题（例如：登录页、计时器、商品卡片列表）
+export const GUIDANCE_SYSTEM_PROMPT = `你是一个提示词写作教练。学生在网页制作课上让 AI 生成或修改网页，你帮他把这一次要对 AI 说的话想清楚、说具体。后面会给你题目要求、他之前对 AI 说过的话和 AI 上一次的说明。
- 至少包含以下一项具体描述：颜色/布局/风格、交互行为（按钮点击、动画等）、页面内容（文字、数量、图标）
+
 判断标准——看学生这一次的话，满足下面一条就算"够好"：
 - 第一次描述作品：除了题目要求，有他自己的想法，至少有一处具体的视觉、交互或内容是他自己定的（如配色风格、玩法规则、页面上放什么）
 - 修改已有网页：说清楚要改哪里、改成什么样；或者说清楚出了什么问题——做了什么操作、想看到什么、实际看到了什么
 判断为"不够好"的典型情况：
- 目标太泛，如"做一个好看的页面"
+- 只有作品名称，如"做一个象棋游戏"
- 只有主题，完全没有任何视觉、交互或内容描述
+- 照搬或改写题目要求，没有自己的想法——哪怕写得很长、功能很多
 - 只说"不行""有 bug""再好看一点"，没说是哪里、要怎样
 规则：
-1. 如果提示词不够好，用 1-2 个启发性问题引导学生补充细节，不要直接给出答案
+1. 如果不够好，用 1-2 个启发性问题引导学生补充，问他自己的想法，不要直接给出答案，也不要列一串功能让他挑
-2. 如果提示词已经够好，以 [READY] 开头回复，简短夸奖学生并说明可以生成了
+2. 照搬题目要求时，先肯定他读懂了题目，再问题目之外他想让作品有什么不一样
-3. 用中文回复，语气鼓励，简洁明了
+3. AI 上一次的说明里如果有“我替你定的”，可以问他同不同意这些决定
-4. 使用 Markdown 语法高亮关键词，优先突出 **主题**、**视觉**、**交互**、**内容**、**可以生成** 等重点
+4. 如果已经够好，以 [READY] 开头回复，简短说明好在哪里，告诉他可以生成了；他还给了理由（为什么这样改）的，特别夸一下
-5. 如果回复以 [READY] 开头，[READY] 不要加粗，必须保持原始文本
+5. 用中文回复，语气鼓励，简洁明了
-6. 不要生成任何代码`
+6. 使用 Markdown 语法高亮关键词，优先突出 **视觉**、**交互**、**内容**、**你的想法**、**可以生成** 等重点
 7. 如果回复以 [READY] 开头，[READY] 不要加粗，必须保持原始文本
 8. 不要生成任何代码`
 // 走火山方舟（ARK）接口的模型
 const ARK_MODELS = new Set(["doubao-seed-2-1-lite-260915"])
@@ -59,10 +83,14 @@ export interface ChatMessage {
  content: string
 }
 // DeepSeek（deepseek-flash）默认先思考再回答，思考的内容在 reasoning_content 里，也算输出 token；
 // 思考模式下 temperature 不起作用。thinking 只发给 DeepSeek，豆包不传
 interface CompletionOptions {
  model?: string
  stream: boolean
  thinking?: boolean
  maxTokens?: number
  temperature?: number
  signal?: AbortSignal
  timeoutMs?: number
 }
@@ -81,7 +109,9 @@ function buildRequest(messages: ChatMessage[], opts: CompletionOptions) {
    messages,
    stream: opts.stream,
  }
  if (!ark && opts.thinking !== undefined) body.thinking = { type: opts.thinking ? "enabled" : "disabled" }
  if (opts.maxTokens) body.max_tokens = opts.maxTokens
  if (opts.temperature !== undefined) body.temperature = opts.temperature
  const apiKey = ark ? config.arkApiKey : config.llmApiKey
  if (!apiKey) throw new Error("未配置模型接口密钥")
  const baseUrl = (ark ? config.arkBaseUrl : config.llmBaseUrl).replace(/\/$/, "")
@@ -107,8 +137,7 @@ async function ensureOk(res: Response) {
 // 读取 OpenAI 兼容接口的 SSE 流，逐段产出文本
 async function* streamCompletion(
  messages: ChatMessage[],
-  model?: string,
+  { model, thinking, signal }: { model?: string; thinking?: boolean; signal?: AbortSignal },
  signal?: AbortSignal,
 ): AsyncGenerator<string> {
  // 带思考的模型生成整页代码可能超过两分钟，只要一直有数据就不该中断
  const idle = new AbortController()
@@ -123,6 +152,7 @@ async function* streamCompletion(
    const res = await buildRequest(messages, {
      model,
      stream: true,
      thinking,
      signal: AbortSignal.any(signals),
      timeoutMs: STREAM_MAX_MS,
    })
@@ -139,6 +169,7 @@ async function* streamCompletion(
        if (data === "[DONE]") return true
        const json = JSON.parse(data)
        if (json.error) throw new Error(json.error.message ?? JSON.stringify(json.error))
        // 思考阶段只有 delta.reasoning_content，不发给学生（它也会重置空闲计时）
        const content = json.choices?.[0]?.delta?.content
        if (content) yield content
      }
@@ -158,14 +189,15 @@ async function* streamCompletion(
  }
 }
-// 发给模型的对话上下文：只留最近 20 条；除最后一条 AI 回复外，旧回复里的代码块换成占位文字。
+// 发给模型的对话上下文：只留最近 20 条；除最后一条带代码的 AI 回复外，旧回复里的代码块换成占位文字。
-// 每次回复都是完整代码，模型只需要最新那份；否则对话越长 token 越多，最后会超出上下文长度
+// 每次回复都是完整代码，模型只需要最新那份；否则对话越长 token 越多，最后会超出上下文长度。
 // 只回答问题的回复没有代码，不能拿它当“最后一条”，不然最新的代码也被省略了
 const MAX_HISTORY = 20
 const CODE_BLOCK = /```[\s\S]*?```/g
 export function trimHistory(history: ChatMessage[]): ChatMessage[] {
  const recent = history.slice(-MAX_HISTORY)
-  const last = recent.findLastIndex((m) => m.role === "assistant")
+  const last = recent.findLastIndex((m) => m.role === "assistant" && m.content.includes("```"))
  return recent.map((m, i) =>
    m.role === "assistant" && i !== last
      ? { ...m, content: m.content.replace(CODE_BLOCK, "```\n（旧版本代码已省略）\n```") }
@@ -185,30 +217,34 @@ export function userFacingError(e: unknown) {
  return status ? `AI 服务出错（${status}），请稍后重试` : "AI 服务出错，请稍后重试"
 }
-export function streamChat(history: ChatMessage[], model?: string, signal?: AbortSignal) {
+// 生成整页代码：开着思考，质量优先。brief 是挑战的题目要求（见 challengeBrief），自由创作没有
-  return streamCompletion(
+export function streamChat(history: ChatMessage[], model?: string, signal?: AbortSignal, brief?: string | null) {
-    [{ role: "system", content: SYSTEM_PROMPT }, ...trimHistory(history)],
+  const system = brief ? `${SYSTEM_PROMPT}\n\n${BRIEF_RULES}\n\n${brief}` : SYSTEM_PROMPT
-    chatModel(model),
+  return streamCompletion([{ role: "system", content: system }, ...trimHistory(history)], {
    model: chatModel(model),
    thinking: true,
    signal,
-  )
+  })
 }
-export function streamGuidance(history: ChatMessage[], signal?: AbortSignal) {
+// 引导只是简短追问，不用思考，回得快。context 是题目和学生之前的对话（见 guidanceContext）
-  return streamCompletion(
+export function streamGuidance(history: ChatMessage[], context: string, signal?: AbortSignal) {
-    [{ role: "system", content: GUIDANCE_SYSTEM_PROMPT }, ...history],
+  return streamCompletion([{ role: "system", content: `${GUIDANCE_SYSTEM_PROMPT}\n\n${context}` }, ...history], {
-    undefined,
+    thinking: false,
    signal,
-  )
+  })
 }
 export async function complete(
  messages: ChatMessage[],
-  opts: { model?: string; maxTokens?: number; timeoutMs?: number } = {},
+  opts: { model?: string; thinking?: boolean; maxTokens?: number; temperature?: number; timeoutMs?: number } = {},
 ): Promise<string> {
  const res = await buildRequest(messages, {
    model: opts.model ?? config.llmModel,
    stream: false,
    thinking: opts.thinking,
    maxTokens: opts.maxTokens,
    temperature: opts.temperature,
    timeoutMs: opts.timeoutMs,
  })
  await ensureOk(res)
@@ -1,5 +1,6 @@
 import { all, get, run, now, newUuidHex } from "../db"
 import { Role, type User } from "../auth"
 import { pyRound } from "./pyround"
 export interface MessageRow {
  id: number
@@ -98,6 +99,62 @@ export function codeAsMarkdown(code: {
  return "```html\n" + parts.join("\n\n") + "\n```"
 }
 // 题目要求去掉图片、压成一段，太长的截掉
 const BRIEF_CHARS = 800
 export function briefOf(content: string, max = BRIEF_CHARS) {
  const text = content
    .replace(/!\[[^\]]*\]\([^)]*\)/g, "")
    .replace(/\s+/g, " ")
    .trim()
  return clipTo(text, max)
 }
 function clipTo(text: string, n: number) {
  return text.length > n ? text.slice(0, n) + "……" : text
 }
 // 挑战的题目要求，给生成网页的 AI 当对照清单（见 llm.ts 的 BRIEF_RULES）；要逐条对照，不能截得太短。
 // 自由创作的题目是学生自己写的介绍，不用对照，返回 null
 export function challengeBrief(taskId: number) {
  const task = get<{ title: string; content: string }>(
    `SELECT title, content FROM task_task WHERE id = $id AND task_type = 'challenge'`,
    { id: taskId },
  )
  const brief = task && briefOf(task.content, 4000)
  return brief ? `题目：${task.title}\n\n题目要求：${brief}` : null
 }
 // “帮我写清楚”的教练要知道题目和学生已经做到哪了，才分得清照搬题目和自己的想法：
 // 题目要求、学生最近几次对 AI 说的话、AI 上一次的说明（代码去掉）
 export function guidanceContext(conversationId: string, taskId: number) {
  const task = get<{ title: string; content: string }>(
    `SELECT title, content FROM task_task WHERE id = $id`,
    { id: taskId },
  )
  const asked = all<{ content: string }>(
    `SELECT content FROM prompt_message
     WHERE conversation_id = $id AND role = 'user' AND source IN ('conversation', 'manual')
     ORDER BY created DESC, id DESC LIMIT 5`,
    { id: conversationId },
  ).reverse()
  const reply = get<{ content: string }>(
    `SELECT content FROM prompt_message
     WHERE conversation_id = $id AND role = 'assistant' AND source = 'conversation'
     ORDER BY created DESC, id DESC LIMIT 1`,
    { id: conversationId },
  )
  const replyText = reply?.content.replace(/```[\s\S]*?(```|$)/g, "").trim()
  return [
    `题目：${task?.title ?? "（未知）"}`,
    `题目要求：${task ? briefOf(task.content) || "（无）" : "（无）"}`,
    `学生之前对 AI 说过的话：${
      asked.length ? asked.map((m, i) => `\n${i + 1}. ${clipTo(m.content.trim(), 300)}`).join("") : "（还没有，这是第一次描述作品）"
    }`,
    `AI 上一次的说明：${replyText ? clipTo(replyText, 600) : "（无）"}`,
  ].join("\n\n")
 }
 export function getPrecedingUserMessage(asst: Pick<MessageRow, "id" | "conversation_id" | "created">) {
  return get<MessageRow>(
    `SELECT * FROM prompt_message
@@ -141,3 +198,38 @@ export function buildPromptRounds(source: MessageRow) {
  }
  return rounds
 }
 // 课堂总结：这些提交各自是怎么问出来的（只走覆盖索引，不读提问和代码）
 export function promptChainStats(submissionIds: string[]) {
  type Slim = { conversation_id: string; created: string; id: number; role: string; source: string; prompt_level: number | null }
  const finals = all<Omit<Slim, "role" | "prompt_level">>(
    `SELECT conversation_id, created, id, source FROM prompt_message
     WHERE submission_id IN (SELECT value FROM json_each($ids))`,
    { ids: JSON.stringify(submissionIds) },
  )
  const byConv = new Map<string, Slim[]>()
  const msgs = all<Slim>(
    `SELECT conversation_id, created, id, role, source, prompt_level FROM prompt_message
     WHERE conversation_id IN (SELECT value FROM json_each($convs))`,
    { convs: JSON.stringify([...new Set(finals.map((f) => f.conversation_id))]) },
  )
  for (const m of msgs) byConv.set(m.conversation_id, [...(byConv.get(m.conversation_id) ?? []), m])
  const out = { with_chain: finals.length, one_round: 0, external: 0, used_guidance: 0, avg_rounds: 0 }
  const levels = [0, 0, 0, 0, 0, 0]
  let rounds = 0
  for (const f of finals) {
    // 和 buildPromptRounds 一样：只算交的那一版之前的提问
    const asked = (byConv.get(f.conversation_id) ?? []).filter(
      (m) => m.role === "user" && (m.created < f.created || (m.created === f.created && m.id < f.id)),
    )
    const code = asked.filter((m) => m.source !== "guidance")
    rounds += code.length
    if (code.length === 1) out.one_round++
    if (f.source === "manual") out.external++
    if (code.length < asked.length) out.used_guidance++
    for (const m of code) if (m.prompt_level) levels[m.prompt_level - 1]!++
  }
  out.avg_rounds = finals.length ? pyRound(rounds / finals.length, 1) : 0
  return { ...out, levels }
 }
@@ -3,8 +3,10 @@ import { all, get, run, iso, uuidOut } from "./db"
 import { userFromAccessToken, type User } from "./auth"
 import {
  canAccessTask,
  challengeBrief,
  codeAsMarkdown,
  getOrCreateActiveConversation,
  guidanceContext,
  insertMessage,
 } from "./services/prompt"
 import {
@@ -15,6 +17,7 @@ import {
  userFacingError,
  type ChatMessage,
 } from "./services/llm"
 import { classifyMessage, inBackground } from "./services/classifier"
 type Kind = "prompt" | "guidance"
@@ -116,11 +119,12 @@ async function handlePrompt(ws: WS, data: any) {
  const model = String(data.model ?? "")
  if (!prompt) return
-  ws.data.pendingUserMessageId = insertMessage({
+  const userMessageId = insertMessage({
    conversation_id: ws.data.conversationId,
    role: "user",
    content: prompt,
  })
  ws.data.pendingUserMessageId = userMessageId
  try {
    const history: ChatMessage[] = conversationHistory(ws.data.conversationId).map((m) => ({
      role: m.role as ChatMessage["role"],
@@ -129,7 +133,8 @@ async function handlePrompt(ws: WS, data: any) {
    }))
    let full: string
    try {
-      full = await streamToClient(ws, streamChat(history, model, ws.data.abort.signal))
+      const brief = challengeBrief(ws.data.taskId)
      full = await streamToClient(ws, streamChat(history, model, ws.data.abort.signal, brief))
    } catch (e) {
      // 客户端自己断开导致的中止不算出错
      if (!ws.data.closed) send(ws, { type: "error", content: userFacingError(e) })
@@ -146,8 +151,12 @@ async function handlePrompt(ws: WS, data: any) {
      code_css: code.css,
      code_js: code.js,
    })
    // 有了回复，这条提问才留下来；在后台打 L1–L6 层级
    inBackground(classifyMessage(userMessageId))
    ws.data.pendingUserMessageId = null
-    send(ws, { type: "complete", code, message_id: messageId })
+    // 只回答问题、没给代码的回复不是新版本；发 null，免得工作台把网页清空
    const hasCode = !!(code.html || code.css || code.js)
    send(ws, { type: "complete", code: hasCode ? code : null, message_id: messageId })
  } finally {
    dropPendingUserMessage(ws)
  }
@@ -157,17 +166,22 @@ async function handleGuidance(ws: WS, data: any) {
  const prompt = String(data.content ?? "").trim()
  if (!prompt) return
-  ws.data.pendingUserMessageId = insertMessage({
+  const userMessageId = insertMessage({
    conversation_id: ws.data.conversationId,
    role: "user",
    content: prompt,
    source: "guidance",
  })
  ws.data.pendingUserMessageId = userMessageId
  ws.data.guidanceHistory.push({ role: "user", content: prompt })
  try {
    let full: string
    try {
-      full = await streamToClient(ws, streamGuidance(ws.data.guidanceHistory, ws.data.abort.signal))
+      const context = guidanceContext(ws.data.conversationId, ws.data.taskId)
      full = await streamToClient(
        ws,
        streamGuidance(ws.data.guidanceHistory, context, ws.data.abort.signal),
      )
    } catch (e) {
      // 这条提问没有回复，留在上下文里会出现连续两条用户消息
      ws.data.guidanceHistory.pop()
@@ -185,6 +199,7 @@ async function handleGuidance(ws: WS, data: any) {
      content: clean,
      source: "guidance",
    })
    inBackground(classifyMessage(userMessageId))
    ws.data.pendingUserMessageId = null
    send(ws, { type: "complete", message_id: messageId, is_ready: isReady })
  } finally {
@@ -276,6 +276,34 @@ describe("提交与删除", () => {
    expect(all(`SELECT COUNT(*) AS n FROM submission_submission WHERE id != $id`, { id: otherSub })).toEqual([{ n: 1 }])
  })
  test("交的时候写的反思：再交没写就保留，只有老师和作者能看，删提交时一起删", async () => {
    makeUser("stu")
    makeUser("mate")
    makeUser("teacher", "admin")
    const task = makeTask("T")
    const stu = await client("stu")
    const edit = await stu("POST", "/api/prompt/versions", { task_id: task, html: "<p>1", css: "", js: "" })
    const sub = await stu("POST", "/api/submission/", {
      task_id: task,
      html: "<p>1",
      message_id: edit.json.message_id,
      reflection: "  按钮改成三个，因为五个太乱  ",
    })
    const id = sub.json.id
    await stu("POST", "/api/submission/", { task_id: task, html: "<p>1!", message_id: edit.json.message_id, reflection: "" })
    expect((await stu("GET", `/api/submission/${id}`)).json.reflection).toBe("按钮改成三个，因为五个太乱")
    expect((await (await client("teacher"))("GET", `/api/submission/${id}`)).json.reflection).toBe("按钮改成三个，因为五个太乱")
    const mate = await client("mate")
    await mate("POST", "/api/submission/", { task_id: task, html: "<p>m" })
    const peer = await mate("GET", `/api/submission/${id}`)
    expect(peer.status).toBe(200)
    expect(peer.json.reflection).toBeNull()
    expect((await stu("DELETE", `/api/submission/${id}`)).status).toBe(200)
    expect(all(`SELECT * FROM submission_reflection`)).toEqual([])
  })
  test("被评分的提交只有超管能删", async () => {
    const author = makeUser("author")
    const other = makeUser("other")
@@ -645,4 +673,50 @@ describe("课堂总结", () => {
    const all = await adm("GET", `/api/submission/stats/${task}`)
    expect(all.json.top_scored.map((t: any) => t.username)).toEqual(["weba2小李", "weba1小明"])
  })
  test("统计每人最新一版的提示词链", async () => {
    const { createConversation, insertMessage } = await import("../src/services/prompt")
    makeUser("adm", "admin")
    const task = makeTask("挑战")
    const say = (conv: string, source: string, level: number | null, submission_id?: string) => {
      const id = insertMessage({ conversation_id: conv, role: "user", source, content: "提问" })
      run(`UPDATE prompt_message SET prompt_level = $level WHERE id = $id`, { level, id })
      insertMessage({ conversation_id: conv, role: "assistant", source, content: "回答", submission_id })
    }
    // 小明：AI 对话 → 引导 → 外部 AI，交最后这版；交了之后的提问不算
    const a = makeUser("weba1小明", "normal", "a1")
    const convA = createConversation(a, task)
    say(convA, "conversation", 3)
    say(convA, "guidance", 2)
    say(convA, "manual", 6, makeSubmission(a, task))
    say(convA, "conversation", 1)
    // 小红：一轮就交
    const b = makeUser("weba1小红", "normal", "a1")
    say(createConversation(b, task), "conversation", 1, makeSubmission(b, task))
    // 小刚：手动交的，没有提示词
    makeSubmission(makeUser("weba1小刚", "normal", "a1"), task)
    const res = await (await client("adm"))("GET", `/api/submission/stats/${task}`)
    expect(res.json.prompts).toEqual({
      with_chain: 2,
      one_round: 1,
      external: 1,
      used_guidance: 1,
      avg_rounds: 1.5,
      levels: [1, 0, 1, 0, 0, 1],
    })
  })
 })
 describe("提示词分类", () => {
  test("太短的提问不打层级，以前打的也清掉", async () => {
    const { classifyMessage } = await import("../src/services/classifier")
    const { createConversation, insertMessage } = await import("../src/services/prompt")
    const conv = createConversation(makeUser("stu"), makeTask("T"))
    const id = insertMessage({ conversation_id: conv, role: "user", content: " 1 " })
    run(`UPDATE prompt_message SET prompt_level = 6 WHERE id = $id`, { id })
    expect(await classifyMessage(id)).toBeNull()
    const row = get<{ prompt_level: number | null }>(`SELECT prompt_level FROM prompt_message WHERE id = $id`, { id })
    expect(row!.prompt_level).toBeNull()
  })
 })
@@ -5,6 +5,7 @@ import {
  extractCode,
  parseGuidanceResponse,
  streamChat,
  streamGuidance,
  trimHistory,
  userFacingError,
 } from "../src/services/llm"
@@ -77,12 +78,13 @@ test("CSV 转义", () => {
 })
 describe("streamChat", () => {
-  test("不认识的模型换成默认模型，最后一行没有换行也能读到", async () => {
+  test("不认识的模型换成默认模型，最后一行没有换行也能读到；生成代码开思考，引导不开", async () => {
    let model = ""
    let thinking: unknown
    const server = Bun.serve({
      port: 0,
      async fetch(req) {
-        model = ((await req.json()) as any).model
+        ;({ model, thinking } = (await req.json()) as any)
        const line = (s: string) => `data: ${JSON.stringify({ choices: [{ delta: { content: s } }] })}`
        // 一行拆在两个分块之间，最后一行不带换行
        const body = `${line("你好")}\n\n${line("世")}\n${line("界")}`
@@ -108,6 +110,9 @@ describe("streamChat", () => {
      for await (const s of streamChat([{ role: "user", content: "hi" }], "deepseek-v4-flash-thinking")) out += s
      expect(out).toBe("你好世界")
      expect(model).toBe(config.llmModel)
      expect(thinking).toEqual({ type: "enabled" })
      for await (const _ of streamGuidance([{ role: "user", content: "hi" }], "题目：x"));
      expect(thinking).toEqual({ type: "disabled" })
    } finally {
      Object.assign(config, saved)
      server.stop(true)
@@ -135,6 +140,19 @@ test("trimHistory 只留最近 20 条，旧回复的代码换成占位", () => {
  expect(out.filter((m) => m.content.includes("<p>"))).toHaveLength(1)
 })
 test("trimHistory 最后一条回复没有代码时，保留之前最新的那份代码", () => {
  const out = trimHistory([
    { role: "user", content: "做个页面" },
    { role: "assistant", content: "第1版\n```html\n<p>1</p>\n```" },
    { role: "user", content: "加个按钮" },
    { role: "assistant", content: "第2版\n```html\n<p>2</p>\n```" },
    { role: "user", content: "为什么按钮是圆的？" },
    { role: "assistant", content: "因为加了 border-radius。" },
  ])
  expect(out[3]!.content).toContain("<p>2</p>")
  expect(out[1]!.content).not.toContain("<p>1</p>")
 })
 test("userFacingError 不透出接口原始返回", () => {
  const quiet = console.error
  console.error = () => {}
@@ -2,18 +2,25 @@ import { afterAll, beforeAll, beforeEach, describe, expect, test } from "bun:tes
 import type { Server } from "bun"
 import { app } from "../src/app"
 import { config } from "../src/config"
-import { all } from "../src/db"
+import { all, run } from "../src/db"
 import { upgrade, websocket, type WsData } from "../src/ws"
 import { client, freshDb, makeTask, makeUser } from "./helpers"
 // 假的模型接口：每段之间停一下，方便测试中途断开
-const PIECES = ["好的", "```html\n<p>hi</p>\n```"]
+const DEFAULT_PIECES = ["好的", "```html\n<p>hi</p>\n```"]
-// 最近一次发给模型的请求
+let PIECES = DEFAULT_PIECES
 // 最近一次发给模型的请求；分类（非流式）单独记，固定回“3”
 let lastRequest: any = null
 let lastClassify: any = null
 const llm = Bun.serve({
  port: 0,
  async fetch(req) {
-    lastRequest = await req.json()
+    const body: any = await req.json()
    if (!body.stream) {
      lastClassify = body
      return Response.json({ choices: [{ message: { content: "3" } }] })
    }
    lastRequest = body
    const enc = new TextEncoder()
    const stream = new ReadableStream({
      async start(ctrl) {
@@ -50,7 +57,10 @@ afterAll(() => {
  llm.stop(true)
 })
-beforeEach(freshDb)
+beforeEach(() => {
  freshDb()
  PIECES = DEFAULT_PIECES
 })
 interface Conn {
  ws: WebSocket
@@ -106,6 +116,19 @@ describe("WebSocket 代码生成", () => {
    c.ws.close()
  })
  test("回复完后在后台给提问打层级", async () => {
    const { task, token } = await setup()
    const c = (await connect(`/ws/prompt/${task}/?token=${token}`))!
    await c.next("init")
    c.ws.send(JSON.stringify({ type: "message", content: "做个页面" }))
    await c.next("complete")
    const level = () => all(`SELECT prompt_level FROM prompt_message WHERE role = 'user'`)[0]?.prompt_level
    for (let i = 0; i < 50 && level() == null; i++) await Bun.sleep(20)
    expect(level()).toBe(3)
    expect(lastClassify.thinking).toEqual({ type: "disabled" })
    c.ws.close()
  })
  test("AI 回复中再发消息会被拒绝", async () => {
    const { task, token } = await setup()
    const c = (await connect(`/ws/prompt/${task}/?token=${token}`))!
@@ -157,9 +180,70 @@ describe("WebSocket 代码生成", () => {
    c.ws.close()
  })
  test("只回答问题、没给代码时 code 是 null", async () => {
    const { task, token } = await setup()
    PIECES = ["因为加了 ", "border-radius。"]
    const c = (await connect(`/ws/prompt/${task}/?token=${token}`))!
    await c.next("init")
    c.ws.send(JSON.stringify({ type: "message", content: "为什么按钮是圆的？" }))
    const done = await c.next("complete")
    expect(done.code).toBeNull()
    expect(done.message_id).toBeGreaterThan(0)
    c.ws.close()
  })
  test("挑战的题目要求作为对照清单发给 AI，自由创作不发", async () => {
    const { task, token } = await setup()
    run(`UPDATE task_task SET content = $c WHERE id = $id`, { c: "做一个计时器，有开始和暂停按钮", id: task })
    const c = (await connect(`/ws/prompt/${task}/?token=${token}`))!
    await c.next("init")
    c.ws.send(JSON.stringify({ type: "message", content: "做个页面" }))
    await c.next("complete")
    expect(lastRequest.messages[0].content).toContain("它是对照清单")
    expect(lastRequest.messages[0].content).toContain("题目要求：做一个计时器，有开始和暂停按钮")
    c.ws.close()
    const creation = makeTask("我的作品", "creation", 2)
    run(`UPDATE task_task SET content = '我想做一个相册' WHERE id = $id`, { id: creation })
    const d = (await connect(`/ws/prompt/${creation}/?token=${token}`))!
    await d.next("init")
    d.ws.send(JSON.stringify({ type: "message", content: "做个页面" }))
    await d.next("complete")
    expect(lastRequest.messages[0].content).not.toContain("题目要求")
    d.ws.close()
  })
  test("token 无效或任务不存在时握手失败", async () => {
    const { task, token } = await setup()
    expect(await connect(`/ws/prompt/${task}/?token=bad`)).toBeNull()
    expect(await connect(`/ws/prompt/9999/?token=${token}`)).toBeNull()
  })
 })
 describe("WebSocket 帮我写清楚", () => {
  test("教练能看到题目要求、之前的提问和 AI 上一次的说明", async () => {
    const { task, token } = await setup()
    run(`UPDATE task_task SET content = $c WHERE id = $id`, { c: "做一个计时器，![图](a.png) 有开始和暂停按钮", id: task })
    const p = (await connect(`/ws/prompt/${task}/?token=${token}`))!
    await p.next("init")
    p.ws.send(JSON.stringify({ type: "message", content: "做个计时器" }))
    await p.next("complete")
    p.ws.close()
    const g = (await connect(`/ws/guidance/${task}/?token=${token}`))!
    await g.next("init")
    g.ws.send(JSON.stringify({ type: "message", content: "改成粉色" }))
    await g.next("complete")
    const system = lastRequest.messages[0]
    expect(system.role).toBe("system")
    expect(system.content).toContain("题目要求：做一个计时器， 有开始和暂停按钮")
    expect(system.content).toContain("1. 做个计时器")
    expect(system.content).toContain("AI 上一次的说明：好的")
    expect(system.content).not.toContain("<p>hi</p>")
    expect(lastRequest.messages.at(-1)).toEqual({ role: "user", content: "改成粉色" })
    // 等后台打完层级再结束，不然假接口关了它会报错
    const level = () => all(`SELECT prompt_level FROM prompt_message WHERE source = 'guidance' AND role = 'user'`)[0]?.prompt_level
    for (let i = 0; i < 50 && level() == null; i++) await Bun.sleep(20)
    g.ws.close()
  })
 })
Author	SHA1	Message	Date
xuyueandClaude Opus 5.5	489274ba0e	feat: 加 prompt-stats 脚本，对比某天前后学生和 AI 对话的情况 Deploy / deploy (build, debian, 22) (push) Canceled after 0s Details 轮数分布、层级分布、第 2 轮起的 L1、只粘贴外部代码的比例、“出问题了”“帮我写清楚” 和交作品反思的使用情况。只读。 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 22:16:33 +08:00
xuyueandClaude Opus 5.5	e9127318cb	feat: “出问题了”三栏说清问题，交作品时写一句反思，外部提示词至少 8 个字 Deploy / deploy (build, debian, 22) (push) Canceled after 0s Details - 对话框加“出问题了”：做了什么 / 想看到什么 / 实际看到了什么拼成一句发给 AI；只说“还是不行”这类的提醒一次，可以直接发 - “交这一版”先弹框问最满意的改动和理由（可以不写），存在 submission_reflection，批改面板显示为“学生说”，只有老师和作者本人看得到 - 粘贴外部 AI 代码时提示词至少 8 个字，可以把几轮都贴进来 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 22:10:48 +08:00
xuyueandClaude Opus 5.5	5bdacc438d	feat: 挑战的题目要求发给 AI 当对照清单，学生没提到的不替他做，在“想一想”里问 Deploy / deploy (build, debian, 22) (push) Canceled after 0s Details 学生不用再整段贴题目；AI 只做学生说了的，从没提到的要求里挑一项问他打算怎么做。整段贴了题目的照做。自由创作的题目是学生自己写的，不发。输入框提示改成“用你自己的话说说”。 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 21:40:04 +08:00
xuyueandClaude Opus 5.5	43c18cdb88	feat: AI 回复说明替学生定了什么并问一个问题，“帮我写清楚”能看到题目和之前的对话 Deploy / deploy (build, debian, 22) (push) Canceled after 0s Details - 生成回复分“做了什么 / 我替你定的 / 想一想”三项；只问问题时直接回答、不给代码 - 没有代码的回复发 code: null，不再清空网页；上下文保留最近一份带代码的回复 - 教练带上题目要求、最近的提问和 AI 上一次的说明，照搬题目不算写清楚 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 21:36:06 +08:00
xuyueandClaude Opus 5.5	9fee688d95	fix: AI 对话和引导里的提问回复完后自动打层级 Deploy / deploy (build, debian, 22) (push) Canceled after 0s Details 旧后端在消息保存后自动分类，改写成 Bun 时漏了：只有外部 AI 版本和手动提交带的提示词会分类，学生在 AI 对话和引导里发的提问都没有层级。现在回复存库后在后台分类（失败或断开的提问不分类）。 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 21:13:46 +08:00
xuyueandClaude Opus 5.5	9f3f7187e1	feat: AI 对话等待时显示构思了多久，代码不在对话里展开 Deploy / deploy (build, debian, 22) (push) Canceled after 0s Details deepseek-flash 生成代码前要思考，复杂的题第一个字要等一两分钟，之前只有三个跳动的点；出字后又把几万字的代码一段段渲染在对话里，学生并不看代码，机房的旧机器也吃力。 - 出字前显示“AI 正在构思… 已 N 秒”，超过 20 秒提示复杂的网页要想一两分钟 - 生成中只显示文字说明和“正在写网页… 已写 N 行”；写完的回复里代码块换成 “网页代码 N 行，在右边‘代码’里看” Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 21:09:47 +08:00
xuyueandClaude Opus 5.5	6b2ffbb3ec	fix: 按 DeepSeek 文档为每种调用指定是否思考，修复分类器拿不到结果 deepseek-flash 默认先思考（reasoning_effort high），思考也占输出 token，思考模式下 temperature 无效。分类器只给 10 个 token，全被思考用掉，正文为空，切到新后端后的提问都没有层级。 - 生成代码：开思考；引导：关思考，回得快；分类：关思考，temperature 0 才生效 - thinking 只发给 DeepSeek，豆包请求不变 - 实测 deepseek-flash：分类约 1.2s，引导约 1.5s；固定 68 条样本两次一致 65 条 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 20:53:30 +08:00
xuyueandClaude Opus 5.5	b765cbf1aa	fix: 理清 L1/L2 的含义，并写死几条容易判错的边界 - L2 理解按布鲁姆的本义改成“想弄懂原理”：问为什么、是什么意思、怎么做到的；只报结果（“没反应”“进不去”“有 bug”）归 L1 - 标准改成按顺序判断，并写明：只说风格和功能没有理由是 L3 不是 L5，说明要改成什么样是 L3 不是 L4，整段贴题目要求是 L3 不是 L1 - 生产库副本固定抽 70 条（含 10 条“为什么”提问），两次结果一致 66/68 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 20:45:44 +08:00
xuyueandClaude Opus 5.5	e06930c8f7	feat: 提示词链显示层级走势、每轮改了多少行；课堂总结加“怎么问的 AI” - 提示词弹框：顶部每轮一根层级柱子，点了跳到那一轮；每轮标出代码比上一版 +/− 多少行（没改标“代码没变”），和前面某轮一字不差的提问标“重复提问” - 课堂总结：每人最新一版平均几轮、一轮就交、最后一版用外部 AI、问过引导的比例，以及各层级提问次数 - prompt_message 加两个覆盖索引，统计只读小列（像素画板全部班级 197 份约 0.3 ms） Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 20:35:33 +08:00
xuyueandClaude Opus 5.5	f41fda3c19	feat: 提示词分类按“指挥 AI 做网页”的场景重写标准，并带上题目和上一轮 - 以前只看单句、例子是学前端知识，把题目要求原样贴过去也判成 L6（首轮近一半 L6）。现在把题目要求和同一对话里上一轮的提问一起发给模型，只看学生自己想的部分：照搬题目功能清单算 L3，重复上一轮不高于上一轮 - L1–L6 说明跟着改成新的标准 - 生产库副本抽 60 条：新标准两次结果一致 57 条；L3 占 75%，L6 从 25 条降到 4 条 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 20:35:33 +08:00
xuyueandClaude Opus 5.5	30630e8125	fix: 提示词链里的引导提问不算轮数，分类器固定结果、跳过过短输入 - 引导对话和 AI 对话存在同一个对话里，提示词链把它当成一轮“AI 对话”，显示“没有网页”。现在标成“引导”，淡化显示，不计入轮数，也不能选中（↑↓ 跳过） - 分类器 temperature 设为 0：同一句提示词以前约两成会被打成不同层级 - 不到 4 个字的提问（“1”“，”等）不打层级，--force 重标时清掉旧标签 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-27 20:26:29 +08:00