ccus-cli 0.2.22 → 0.2.24

package/README.md

@@ -18,6 +18,14 @@
 >
 > Claude **桌面版** 和 **VS Code 插件** 都不支持 statusLine，因此不会调用 `ccus statusline emit`，也就采集不到使用率数据。
 
+## 面板周评分
+
+个人和聚合面板均提供「周评分」：默认 `50 × √(周消息数 / 200) + 50 × √(周累计额度百分点 / 70)`，消息和额度均合计 Claude + Codex。两项达到基准得 100 分，分数不封顶。
+
+面板只显示评分结果。权重和基准集中在 `src/lib/weekly-score.ts` 的 `DEFAULT_SCORE_SETTINGS` 修改：`messageWeight` 为消息占比（额度占比为 100 减去消息占比），`messageBaseline` 为周消息基准，`quotaBaseline` 为周额度基准。修改后重新构建并重启面板服务。
+
+评分按当前查询范围逐周计算，未满一周不折算；查看本周评分请使用 `ccus dashboard serve --range this-week`。聚合面板在「多人对比」中显示周评分，多周取逐周评分的平均值。额度缺失显示 `--`（消息占比为 100% 时可单独评分）。评分仅在页面展示，不改变 bundle 或 CSV。
+
 ## 安装
 
 全局安装（statusline 每次渲染都会调用，推荐全局装好，避免 `npx` 的启动开销）：
@@ -176,10 +184,15 @@ ccus aggregate serve --input-dir ./team-exports
 - 当前导出 bundle / weeklySummary 的 `schemaVersion` 为 `12`。两处 `range.dayStart` 记录统计日开始时间（如 `07:00`），供聚合沿用导出口径。顶层 `pricing` 记录本地价格目录版本、USD 币种和 `event-time-standard-api` 基准；`weeklySummary.apiEquivalentCost` 与 `dailySummaries[].apiEquivalentCost` 分别包含 `claude`、`codex`、`total` 的金额及定价覆盖度。v10 延续 v9 的 Codex token 口径：`inputTokens = max(0, input_tokens - cached_input_tokens)`，缓存输入由 `cacheReadInputTokens` 单列
 - v11 在有未定价请求的成本结果中增加 `unpricedModels: [{ model, requestCount }]`，按原始模型名累计请求数，模型缺失记为 `null`，无未定价请求时省略。该明细仅写入 JSON，不增加页面或 CSV 字段；v10 导出仍可读取，但没有模型明细。
 - 等效 API 成本按每次请求的模型、发生时间和 token 分类套用内置标准同步 API 价格。Claude 区分输入、输出、缓存读取及 5 分钟/1 小时缓存写入；Codex 区分净输入、缓存输入和输出。它不是 Claude/Codex 订阅账单，也不含税费、折扣、批处理、工具附加费或区域溢价
-- 未知模型不会丢弃：有部分请求可定价时金额是已知小计，页面显示“至少为”；全部请求均无法定价时金额为不可用。个人 dashboard 在顶部第五张卡显示 Claude 与 Codex 的合计成本，不显示来源分项。全部模型价格集中维护在 `src/lib/api-pricing-catalog.json`，价格目录随 ccus 发布，统计过程不联网；个人与多人 dashboard 的合计成本卡下链接独立 `pricing.html` 并在新页面打开，价格页只保留一套标题，Codex 排在 Claude 前并按模型版本从新到旧排列。当前 Claude 目录包含 Opus 4.7、Opus 4.8、Opus 5、Sonnet 5 和 Fable 5；Sonnet 5 按事件时间区分活动价与标准价
-- GPT-6 使用 `gpt-6-astra` 定价，兼容 `gpt-6` 别名及推理强度后缀。自 2026-09-03 起，每百万 token 输入 / 缓存读取 / 输出为 $10 / $1 / $50；输入与缓存读取合计超过 272K 时为 $20 / $2 / $75。来源：[OpenAI 模型定价](https://developers.openai.com/api/docs/models/gpt-6-astra)。价格目录版本为 `2026-09-23`
+- 未知模型不会丢弃：有部分请求可定价时金额是已知小计，页面显示“至少为”；全部请求均无法定价时金额为不可用。个人 dashboard 在顶部第五张卡显示 Claude 与 Codex 的合计成本，不显示来源分项。全部模型价格集中维护在 `src/lib/api-pricing-catalog.json`，价格目录随 ccus 发布，统计过程不联网；个人与多人 dashboard 的合计成本卡下链接独立 `pricing.html` 并在新页面打开，价格页只保留一套标题，Codex 排在 Claude 前并按模型版本从新到旧排列。当前 Claude 目录包含 Opus / Sonnet / Haiku 5.5、Fable 5.1 与 Mythos 5 / 5.1，并保留历史模型价格
+- GPT-6 使用 `gpt-6-astra` 定价，兼容 `gpt-6` 别名及推理强度后缀。自 2026-09-03 起，每百万 token 输入 / 缓存读取 / 输出为 $10 / $1 / $50；输入与缓存读取合计超过 272K 时为 $20 / $2 / $75。来源：[OpenAI 模型定价](https://developers.openai.com/api/docs/models/gpt-6-astra)。价格目录版本为 `2026-10-08`
 - 自 2026-09-22 起，`gpt-6-sol` 每百万 token 输入 / 缓存读取 / 输出为 $2 / $0.20 / $10，`gpt-6-luna` 为 $0.10 / $0.01 / $0.50；输入与缓存读取合计超过 272K 时，分别为 $4 / $0.40 / $15 和 $0.20 / $0.02 / $0.75。均兼容推理强度后缀。来源：[OpenAI 价格表](https://developers.openai.com/api/docs/pricing)。
+- 自 2026-09-29 起，`gpt-6.1-sol` 每百万 token 输入 / 缓存读取 / 输出为 $2 / $0.10 / $10；输入与缓存读取合计超过 272K 时为 $4 / $0.20 / $15。兼容推理强度后缀。来源：[OpenAI 模型定价](https://developers.openai.com/api/docs/models/gpt-6.1-sol)。
 - 自 2026-09-22 起，`claude-opus-5-5`（目录名 `claude-opus-5.5`）每百万 token 输入 / 输出 / 缓存读取 / 5 分钟缓存写入 / 1 小时缓存写入为 $4 / $20 / $0.20 / $5 / $8。来源：[官方发布公告](https://www.anthropic.com/claude-opus-5-5)、[缓存计价说明](https://claude.com/blog/what-a-task-costs-on-opus-5-5)。
+- 自 2026-09-01 起，Fable 5.1 与 Mythos 5.1 每百万 token 输入 / 输出 / 缓存读取 / 5 分钟缓存写入 / 1 小时缓存写入为 $10 / $50 / $0.25 / $12.50 / $20；Mythos 5 自 2026-06-09 起为 $10 / $50 / $1 / $12.50 / $20。Mythos 访问受限，但公开价格可用于成本估算。来源：[Claude 价格表](https://platform.claude.com/docs/en/about-claude/pricing)。
+- Sonnet 5.5 自 2026-09-28 起为 $2 / $10 / $0.20 / $2.50 / $4，2026-10-07 起缓存读取降为 $0.10，其余不变。Haiku 5.5 自 2026-10-07 起为 $0.10 / $0.50 / $0.01 / $0.125 / $0.20；全部输入类 token 合计超过 100K 时为 $0.50 / $2.50 / $0.05 / $0.625 / $1。上述金额依次为每百万 token 输入 / 输出 / 缓存读取 / 5 分钟缓存写入 / 1 小时缓存写入。来源：[Claude 价格表](https://platform.claude.com/docs/en/about-claude/pricing)、[发布记录](https://platform.claude.com/docs/en/release-notes/overview)。
+- Sonnet 5 的 $2 / $10 输入 / 输出价格已转为长期标准价，原定 2026-09-01 涨价取消；Sonnet 4.6 自 2026-03-13 起取消长上下文加价，此前事件保留历史计价。来源：[Claude 发布记录](https://platform.claude.com/docs/en/release-notes/overview)。
+- `gpt-5.4-mini` 自 2026-03-17 起，每百万 token 输入 / 缓存读取 / 输出为 $0.75 / $0.075 / $4.50，不设长上下文加价。来源：[OpenAI 模型定价](https://developers.openai.com/api/docs/models/gpt-5.4-mini)。
 - `gpt-5.3-codex-spark` 自 2026-02-12 起按 `gpt-5.3-codex` 价格估算：每百万 token 输入 $1.75、输出 $14、缓存读取 $0.175。这是本项目采用的估算单价，并非 Spark 官方公开 API 定价。
 - 默认文件名会带 git email 的帐号名前缀和起止日期，例如：`alice_export_2026-05-26_to_2026-06-01.json.gz`
 - `userMessageCount` 来自 `~/.claude/projects/**/*.jsonl` 的非 meta `type:user` 事件

package/dist/lib/aggregate-dashboard.js

@@ -8,6 +8,7 @@ const chart_assets_1 = require("./chart-assets");
 const time_1 = require("./time");
 const api_equivalent_cost_1 = require("./api-equivalent-cost");
 const api_pricing_table_1 = require("./api-pricing-table");
+const weekly_score_1 = require("./weekly-score");
 /** 折线 / 图例统一的调色板，保证同一个人在不同图表里颜色一致。 */
 const CHART_PALETTE = ["#5eead4", "#f59e0b", "#a855f7", "#22c55e", "#f87171", "#60a5fa", "#fbbf24", "#34d399"];
 /**
@@ -415,7 +416,7 @@ function renderFiveHourUsageChart(people, detailRows) {
   `;
 }
 /** 按人渲染汇总表，给出消息、token、成本和 usage 核心指标。 */
-function renderPeopleLeaderboard(people) {
+function renderPeopleLeaderboard(people, weeklyRows) {
     if (people.length === 0) {
         return `
       <section class="panel table-panel">
@@ -429,19 +430,30 @@ function renderPeopleLeaderboard(people) {
       </section>
     `;
     }
+    const scores = new Map();
+    for (const row of weeklyRows) {
+        const values = scores.get(row.personKey) ?? [];
+        values.push((0, weekly_score_1.calculateWeeklyScore)(row.userMessageCount, row.sevenDayCumulativeUsagePct, weekly_score_1.DEFAULT_SCORE_SETTINGS));
+        scores.set(row.personKey, values);
+    }
+    const scoreText = (personKey) => {
+        const values = scores.get(personKey) ?? [];
+        if (values.length === 0 || values.some(value => value === null))
+            return "--";
+        return (values.reduce((sum, value) => sum + (value ?? 0), 0) / values.length).toFixed(1);
+    };
     const rows = people
         .map((person, index) => `
       <tr>
         <td class="rank">${index + 1}</td>
         <td><strong>${escapeHtml(person.personKey)}</strong></td>
+        <td><strong>${scoreText(person.personKey)}</strong></td>
         <td>${formatNumber(person.userMessageCount)}</td>
         <td>${formatTokensM(person.inputTokens)}</td>
         <td>${formatTokensM(person.outputTokens)}</td>
         <td>${formatTokensM(person.cacheReadInputTokens)}</td>
         <td>${escapeHtml(costDisplay(person.estimatedApiEquivalentCostUsd, person.pricedApiRequestCount, person.unpricedApiRequestCount))}</td>
         <td>${escapeHtml(person.pricingCatalogVersion ?? "--")}</td>
-        <td>${escapeHtml(statValue(person.fiveHourPeakUsagePct))}</td>
-        <td>${escapeHtml(statValue(person.fiveHourLatestUsagePct))}</td>
         <td>${escapeHtml(statValue(person.sevenDayPeakUsagePct))}</td>
         <td>${escapeHtml(statValue(person.sevenDayLatestUsagePct))}</td>
         <td>${escapeHtml(statValue(person.sevenDayCumulativeUsagePct))}</td>
@@ -455,7 +467,7 @@ function renderPeopleLeaderboard(people) {
           <p class="eyebrow">People</p>
           <h2>多人对比</h2>
         </div>
-        <p class="muted">按用户消息数降序排列，所有数字直接来自 daily/weekly 汇总。</p>
+        <p class="muted">按用户消息数降序排列；评分按周计算，多周取平均，缺失额度显示 --。</p>
       </div>
       <div class="table-wrap">
         <table>
@@ -463,14 +475,13 @@ function renderPeopleLeaderboard(people) {
             <tr>
               <th>#</th>
               <th>personKey</th>
+              <th>周评分</th>
               <th>消息</th>
               <th>Input tokens</th>
               <th>Output tokens</th>
               <th>Cache read tokens</th>
               <th>等效 API 成本</th>
               <th>价格目录</th>
-              <th>5h Peak</th>
-              <th>5h Latest</th>
               <th>7d Peak</th>
               <th>7d Latest</th>
               <th>7d 累计</th>
@@ -798,7 +809,7 @@ function buildAggregateDashboardHtml(detailRows, dailyRows, weeklyRows, generate
           <a class="pricing-link" href="${api_pricing_table_1.API_PRICING_PAGE_FILE}" target="_blank" rel="noopener noreferrer">查看当前价格表</a>
         </article>
       </section>
-      ${renderPeopleLeaderboard(people)}
+      ${renderPeopleLeaderboard(people, weeklyRows)}
       ${renderSevenDayCumulativeChart(people)}
       ${renderFiveHourUsageChart(people, detailRows)}
       ${renderDailyUserRequestChart(people, dailyIndex, dateAxis)}

package/dist/lib/api-equivalent-cost.js

@@ -85,15 +85,15 @@ function normalizeClaudeModel(model) {
         .replaceAll("_", "-")
         .replace(/\[1m\]$/, "")
         .replace(/-thinking(?=-\d{8}$|$)/, "");
-    const canonicalMinor = /^claude-(opus|sonnet|haiku|fable)-(\d+)\.(\d+)(?:-\d{8})?$/.exec(value);
+    const canonicalMinor = /^claude-(opus|sonnet|haiku|fable|mythos)-(\d+)\.(\d+)(?:-\d{8})?$/.exec(value);
     if (canonicalMinor) {
         return `claude-${canonicalMinor[1]}-${canonicalMinor[2]}.${canonicalMinor[3]}`;
     }
-    const modern = /^claude-(opus|sonnet|haiku|fable)-(\d+)-(\d+)(?:-\d{8})?$/.exec(value);
+    const modern = /^claude-(opus|sonnet|haiku|fable|mythos)-(\d+)-(\d+)(?:-\d{8})?$/.exec(value);
     if (modern) {
         return `claude-${modern[1]}-${modern[2]}.${modern[3]}`;
     }
-    const modernMajor = /^claude-(opus|sonnet|haiku|fable)-(\d+)(?:-\d{8})?$/.exec(value);
+    const modernMajor = /^claude-(opus|sonnet|haiku|fable|mythos)-(\d+)(?:-\d{8})?$/.exec(value);
     if (modernMajor) {
         return `claude-${modernMajor[1]}-${modernMajor[2]}`;
     }

package/dist/lib/api-pricing-catalog.json

@@ -1,5 +1,5 @@
 {
-    "catalogVersion": "2026-09-23",
+    "catalogVersion": "2026-10-08",
     "currency": "USD",
     "basis": "event-time-standard-api",
     "entries": [
@@ -99,13 +99,20 @@
             "provider": "claude",
             "model": "claude-sonnet-4.6",
             "effectiveFrom": "2026-02-17T00:00:00.000Z",
-            "effectiveTo": null,
+            "effectiveTo": "2026-03-13T00:00:00.000Z",
             "prices": { "inputUsdPerMillion": 3, "outputUsdPerMillion": 15, "cacheReadInputUsdPerMillion": 0.3, "cacheWrite5mInputUsdPerMillion": 3.75, "cacheWrite1hInputUsdPerMillion": 6 },
             "longContext": {
                 "thresholdInputTokens": 200000,
                 "prices": { "inputUsdPerMillion": 6, "outputUsdPerMillion": 22.5, "cacheReadInputUsdPerMillion": 0.6, "cacheWrite5mInputUsdPerMillion": 7.5, "cacheWrite1hInputUsdPerMillion": 12 }
             }
         },
+        {
+            "provider": "claude",
+            "model": "claude-sonnet-4.6",
+            "effectiveFrom": "2026-03-13T00:00:00.000Z",
+            "effectiveTo": null,
+            "prices": { "inputUsdPerMillion": 3, "outputUsdPerMillion": 15, "cacheReadInputUsdPerMillion": 0.3, "cacheWrite5mInputUsdPerMillion": 3.75, "cacheWrite1hInputUsdPerMillion": 6 }
+        },
         {
             "provider": "claude",
             "model": "claude-opus-4.7",
@@ -131,15 +138,29 @@
             "provider": "claude",
             "model": "claude-sonnet-5",
             "effectiveFrom": "2026-06-30T00:00:00.000Z",
-            "effectiveTo": "2026-09-01T00:00:00.000Z",
+            "effectiveTo": null,
             "prices": { "inputUsdPerMillion": 2, "outputUsdPerMillion": 10, "cacheReadInputUsdPerMillion": 0.2, "cacheWrite5mInputUsdPerMillion": 2.5, "cacheWrite1hInputUsdPerMillion": 4 }
         },
         {
             "provider": "claude",
-            "model": "claude-sonnet-5",
+            "model": "claude-fable-5.1",
             "effectiveFrom": "2026-09-01T00:00:00.000Z",
             "effectiveTo": null,
-            "prices": { "inputUsdPerMillion": 3, "outputUsdPerMillion": 15, "cacheReadInputUsdPerMillion": 0.3, "cacheWrite5mInputUsdPerMillion": 3.75, "cacheWrite1hInputUsdPerMillion": 6 }
+            "prices": { "inputUsdPerMillion": 10, "outputUsdPerMillion": 50, "cacheReadInputUsdPerMillion": 0.25, "cacheWrite5mInputUsdPerMillion": 12.5, "cacheWrite1hInputUsdPerMillion": 20 }
+        },
+        {
+            "provider": "claude",
+            "model": "claude-mythos-5",
+            "effectiveFrom": "2026-06-09T00:00:00.000Z",
+            "effectiveTo": null,
+            "prices": { "inputUsdPerMillion": 10, "outputUsdPerMillion": 50, "cacheReadInputUsdPerMillion": 1, "cacheWrite5mInputUsdPerMillion": 12.5, "cacheWrite1hInputUsdPerMillion": 20 }
+        },
+        {
+            "provider": "claude",
+            "model": "claude-mythos-5.1",
+            "effectiveFrom": "2026-09-01T00:00:00.000Z",
+            "effectiveTo": null,
+            "prices": { "inputUsdPerMillion": 10, "outputUsdPerMillion": 50, "cacheReadInputUsdPerMillion": 0.25, "cacheWrite5mInputUsdPerMillion": 12.5, "cacheWrite1hInputUsdPerMillion": 20 }
         },
         {
             "provider": "claude",
@@ -155,6 +176,31 @@
             "effectiveTo": null,
             "prices": { "inputUsdPerMillion": 4, "outputUsdPerMillion": 20, "cacheReadInputUsdPerMillion": 0.2, "cacheWrite5mInputUsdPerMillion": 5, "cacheWrite1hInputUsdPerMillion": 8 }
         },
+        {
+            "provider": "claude",
+            "model": "claude-sonnet-5.5",
+            "effectiveFrom": "2026-09-28T00:00:00.000Z",
+            "effectiveTo": "2026-10-07T00:00:00.000Z",
+            "prices": { "inputUsdPerMillion": 2, "outputUsdPerMillion": 10, "cacheReadInputUsdPerMillion": 0.2, "cacheWrite5mInputUsdPerMillion": 2.5, "cacheWrite1hInputUsdPerMillion": 4 }
+        },
+        {
+            "provider": "claude",
+            "model": "claude-sonnet-5.5",
+            "effectiveFrom": "2026-10-07T00:00:00.000Z",
+            "effectiveTo": null,
+            "prices": { "inputUsdPerMillion": 2, "outputUsdPerMillion": 10, "cacheReadInputUsdPerMillion": 0.1, "cacheWrite5mInputUsdPerMillion": 2.5, "cacheWrite1hInputUsdPerMillion": 4 }
+        },
+        {
+            "provider": "claude",
+            "model": "claude-haiku-5.5",
+            "effectiveFrom": "2026-10-07T00:00:00.000Z",
+            "effectiveTo": null,
+            "prices": { "inputUsdPerMillion": 0.1, "outputUsdPerMillion": 0.5, "cacheReadInputUsdPerMillion": 0.01, "cacheWrite5mInputUsdPerMillion": 0.125, "cacheWrite1hInputUsdPerMillion": 0.2 },
+            "longContext": {
+                "thresholdInputTokens": 100000,
+                "prices": { "inputUsdPerMillion": 0.5, "outputUsdPerMillion": 2.5, "cacheReadInputUsdPerMillion": 0.05, "cacheWrite5mInputUsdPerMillion": 0.625, "cacheWrite1hInputUsdPerMillion": 1 }
+            }
+        },
         {
             "provider": "codex",
             "model": "gpt-5",
@@ -236,6 +282,13 @@
                 "prices": { "inputUsdPerMillion": 5, "outputUsdPerMillion": 22.5, "cacheReadInputUsdPerMillion": 0.5 }
             }
         },
+        {
+            "provider": "codex",
+            "model": "gpt-5.4-mini",
+            "effectiveFrom": "2026-03-17T00:00:00.000Z",
+            "effectiveTo": null,
+            "prices": { "inputUsdPerMillion": 0.75, "outputUsdPerMillion": 4.5, "cacheReadInputUsdPerMillion": 0.075 }
+        },
         {
             "provider": "codex",
             "model": "gpt-5.5",
@@ -302,6 +355,17 @@
                 "prices": { "inputUsdPerMillion": 4, "outputUsdPerMillion": 15, "cacheReadInputUsdPerMillion": 0.4 }
             }
         },
+        {
+            "provider": "codex",
+            "model": "gpt-6.1-sol",
+            "effectiveFrom": "2026-09-29T00:00:00.000Z",
+            "effectiveTo": null,
+            "prices": { "inputUsdPerMillion": 2, "outputUsdPerMillion": 10, "cacheReadInputUsdPerMillion": 0.1 },
+            "longContext": {
+                "thresholdInputTokens": 272000,
+                "prices": { "inputUsdPerMillion": 4, "outputUsdPerMillion": 15, "cacheReadInputUsdPerMillion": 0.2 }
+            }
+        },
         {
             "provider": "codex",
             "model": "gpt-6-luna",

package/dist/lib/dashboard.js

@@ -1,5 +1,6 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
+exports.buildPersonalWeeklyScores = buildPersonalWeeklyScores;
 exports.summarizeEvents = summarizeEvents;
 exports.bucketizeEvents = bucketizeEvents;
 exports.buildDashboardHtml = buildDashboardHtml;
@@ -8,6 +9,28 @@ const chart_assets_1 = require("./chart-assets");
 const api_pricing_table_1 = require("./api-pricing-table");
 const payload_1 = require("./payload");
 const time_1 = require("./time");
+const weekly_score_1 = require("./weekly-score");
+function buildPersonalWeeklyScores(events, messages, dayStart) {
+    const dayStartMinutes = (0, time_1.parseDayStart)(dayStart);
+    const weekOf = (date, boundary) => (0, time_1.localDateKey)((0, time_1.resolveRange)("this-week", date, boundary).start);
+    const weeks = new Map();
+    const getWeek = (week) => {
+        if (!weeks.has(week))
+            weeks.set(week, { week, personKey: "本人", userMessageCount: 0, sevenDayCumulativeUsagePct: null });
+        return weeks.get(week);
+    };
+    for (const point of messages) {
+        getWeek(weekOf(new Date(`${point.date}T12:00:00`), 0)).userMessageCount += point.userMessageCount + (point.codexUserMessageCount ?? 0);
+    }
+    for (const event of events)
+        getWeek(weekOf(new Date(event.timestamp), dayStartMinutes));
+    const curves = [false, true].map(codex => (0, aggregate_1.buildSevenDayCurveFromEvents)(events.filter(event => (0, payload_1.isCodexSourceEvent)(event) === codex)));
+    for (const row of weeks.values()) {
+        const totals = curves.map(curve => (0, aggregate_1.computeCumulativeSevenDay)(curve.filter(event => weekOf(new Date(event.timestamp), dayStartMinutes) === row.week)));
+        row.sevenDayCumulativeUsagePct = (0, aggregate_1.addNullable)(totals[0], totals[1]);
+    }
+    return [...weeks.values()];
+}
 /** 所有插入到 HTML/SVG 的动态文本都先转义，避免本地页面被注入内容。 */
 function escapeHtml(value) {
     return value
@@ -142,7 +165,7 @@ function renderChart(buckets) {
       <div class="panel-header">
         <div>
           <p class="eyebrow">Recent Trend</p>
-          <h2>Claude 使用率趋势</h2>
+          <h2>使用率趋势</h2>
         </div>
         <p class="muted">按采样时间聚合：5h（实线）来自 rate_limits.five_hour，7d（虚线）来自 rate_limits.seven_day；紫色「7d 分区叠加累计」走右侧 Y 轴，是把 7d 锯齿波还原成的累计真实使用量（口径同团队看板）</p>
       </div>
@@ -300,6 +323,11 @@ function buildDashboardHtml(events, rangeLabel, start, end, dailyUserMessages =
     const totalUserMessages = dailyUserMessages.reduce((sum, point) => sum + point.userMessageCount, 0);
     const codexTotalUserMessages = dailyUserMessages.reduce((sum, point) => sum + (point.codexUserMessageCount ?? 0), 0);
     const combinedUserMessages = totalUserMessages + codexTotalUserMessages;
+    const scoreWeek = (0, time_1.localDateKey)((0, time_1.resolveRange)("this-week", end, (0, time_1.parseDayStart)(dayStart)).start);
+    const scoreRow = buildPersonalWeeklyScores(events, dailyUserMessages, dayStart).find(row => row.week === scoreWeek);
+    const weeklyScore = scoreRow
+        ? (0, weekly_score_1.calculateWeeklyScore)(scoreRow.userMessageCount, scoreRow.sevenDayCumulativeUsagePct, weekly_score_1.DEFAULT_SCORE_SETTINGS)
+        : null;
     return `<!doctype html>
 <html lang="zh-CN">
   <head>
@@ -477,15 +505,10 @@ function buildDashboardHtml(events, rangeLabel, start, end, dailyUserMessages =
         </div>
       </section>
       <section class="stats">
-        <article class="panel stat-card">
-          <h2>Latest 5h usage</h2>
-          <p class="stat-value">${escapeHtml(statValue(summary.fiveHourLatestUsagePct))}</p>
-          <p class="stat-note">最后一条有效 5 小时 usage 样本</p>
-        </article>
-        <article class="panel stat-card">
-          <h2>Peak 5h usage</h2>
-          <p class="stat-value">${escapeHtml(statValue(summary.fiveHourPeakUsagePct))}</p>
-          <p class="stat-note">窗口内观测到的 5 小时使用率峰值</p>
+        <article class="panel stat-card" id="weekly-score">
+          <h2>周评分</h2>
+          <p class="stat-value">${weeklyScore?.toFixed(1) ?? "--"}</p>
+          <p class="stat-note">${scoreWeek} 当周 · 当前查询范围内数据</p>
         </article>
         <article class="panel stat-card">
           <h2>7d 分区叠加累计</h2>

package/dist/lib/weekly-score.js

@@ -0,0 +1,42 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.DEFAULT_SCORE_SETTINGS = void 0;
+exports.calculateWeeklyScore = calculateWeeklyScore;
+exports.renderWeeklyScores = renderWeeklyScores;
+exports.DEFAULT_SCORE_SETTINGS = {
+    messageWeight: 50,
+    messageBaseline: 200,
+    quotaBaseline: 70,
+};
+function calculateWeeklyScore(messages, quota, settings) {
+    const { messageWeight, messageBaseline, quotaBaseline } = settings;
+    if (!Number.isFinite(messageWeight) || messageWeight < 0 || messageWeight > 100
+        || !Number.isFinite(messageBaseline) || messageBaseline <= 0
+        || !Number.isFinite(quotaBaseline) || quotaBaseline <= 0
+        || !Number.isFinite(messages) || messages < 0
+        || (quota !== null && (!Number.isFinite(quota) || quota < 0)))
+        return null;
+    if (quota === null && messageWeight < 100)
+        return null;
+    return messageWeight * Math.sqrt(messages / messageBaseline)
+        + (100 - messageWeight) * Math.sqrt((quota ?? 0) / quotaBaseline);
+}
+function escapeHtml(value) {
+    return value.replaceAll("&", "&amp;").replaceAll("<", "&lt;").replaceAll(">", "&gt;")
+        .replaceAll('"', "&quot;").replaceAll("'", "&#39;");
+}
+/** 两类面板共用评分；权重和基准统一在 DEFAULT_SCORE_SETTINGS 修改。 */
+function renderWeeklyScores(rows) {
+    const settings = exports.DEFAULT_SCORE_SETTINGS;
+    const body = [...rows].sort((a, b) => b.week.localeCompare(a.week) || a.personKey.localeCompare(b.personKey)).map((row) => {
+        const score = calculateWeeklyScore(row.userMessageCount, row.sevenDayCumulativeUsagePct, settings);
+        return `<tr><td>${escapeHtml(row.week)}</td><td>${escapeHtml(row.personKey)}</td>
+      <td><strong>${score?.toFixed(1) ?? "--"}</strong></td></tr>`;
+    }).join("");
+    return `<section class="panel table-panel" id="weekly-score">
+    <h2>周评分</h2>
+    <div class="table-wrap"><table><thead><tr><th>周起始</th><th>用户</th><th>评分</th></tr></thead>
+    <tbody>${body || '<tr><td colspan="3">暂无评分数据</td></tr>'}</tbody></table></div>
+  </section>`;
+}
+//# sourceMappingURL=weekly-score.js.map

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ccus-cli",
-  "version": "0.2.22",
+  "version": "0.2.24",
   "description": "Claude Code statusline usage logger and dashboard CLI",
   "type": "commonjs",
   "bin": {
@@ -17,7 +17,7 @@
     "build": "tsc -p tsconfig.json && node scripts/copy-vendor.js",
     "prepublishOnly": "npm run build",
     "test": "node --test --test-concurrency 1 \"dist/test/**/*.test.js\"",
-    "test:src": "node --import tsx --test --test-concurrency 1 src/test/payload.test.ts src/test/dashboard.test.ts src/test/export.test.ts src/test/storage.test.ts src/test/api-equivalent-cost.test.ts src/test/claude.test.ts src/test/aggregate.test.ts src/test/aggregate-dashboard.test.ts src/test/install.test.ts src/test/debug.test.ts src/test/time.test.ts src/test/day-start.test.ts src/test/update-check.test.ts src/test/sync.test.ts src/test/scheduler.test.ts src/test/api-mode.test.ts src/test/codex-fetcher.test.ts src/test/codex-sessions.test.ts src/test/sessions.test.ts src/test/codex-install.test.ts src/test/cli-codex-notify.test.ts src/test/cli-codex-hook.test.ts src/test/cli-install-codex.test.ts",
+    "test:src": "node --import tsx --test --test-concurrency 1 src/test/payload.test.ts src/test/dashboard.test.ts src/test/weekly-score.test.ts src/test/export.test.ts src/test/storage.test.ts src/test/api-equivalent-cost.test.ts src/test/claude.test.ts src/test/aggregate.test.ts src/test/aggregate-dashboard.test.ts src/test/install.test.ts src/test/debug.test.ts src/test/time.test.ts src/test/day-start.test.ts src/test/update-check.test.ts src/test/sync.test.ts src/test/scheduler.test.ts src/test/api-mode.test.ts src/test/codex-fetcher.test.ts src/test/codex-sessions.test.ts src/test/sessions.test.ts src/test/codex-install.test.ts src/test/cli-codex-notify.test.ts src/test/cli-codex-hook.test.ts src/test/cli-install-codex.test.ts",
     "smoke:api": "node scripts/smoke-api.mjs"
   },
   "keywords": [