SKILL.md

@@ -1,97 +1,75 @@
 ---
 name: 1688-logistics-scraper
-description: "Scrape 1688 product pages via Chrome, capture full-page screenshots and detail images for vision-based extraction of weight/size data. Use when the user provides a 1688 product URL and needs logistics specs."
+description: "Extract product weight/size/logistics data from 1688 product pages via Chrome browser, output structured JSON. Use when the user provides a 1688 product URL and needs logistics specs."
 ---
 
 # 1688 Logistics Scraper
 
-Capture 1688 product pages and extract weight/size data via vision.
+Extract product weight, size, and logistics data from 1688 product pages.
 
 ## Run
 
 ```bash
-bun scripts/run.ts scrape <url> [--dry-run] [--port=9222]
+bun scripts/run.ts scrape <url> [--dry-run]
+```
+
+### Examples
+
+```bash
+bun scripts/run.ts scrape 'https://detail.1688.com/offer/852504650877.html'
+bun scripts/run.ts scrape 'https://detail.1688.com/offer/852504650877.html' --dry-run
 ```
 
 ## What It Does
 
-1. Opens the 1688 product URL in the browser (default port 18800)
+1. Opens the 1688 product URL in the browser
-2. Scrolls through the entire page, capturing full-page screenshots
+2. Extracts weight/size data from wherever it appears on the page — product attributes, variant specs, logistics section
-3. Downloads all product detail images
+3. Downloads detail images (商品详情图片) for analysis — weight/size is often only in images
-4. Saves to `/tmp/1688-logistics/<offer-id>/`
+4. Outputs structured JSON
 
-## After Running — MUST follow
+## Where To Look For Data
 
-Read ALL screenshots and detail images, then output the following JSON structure. This is the final output for API consumption.
+Weight/size data on 1688 pages hides in multiple places. Check all before giving up:
+
+1. **Product attributes** (商品属性 / 商品参数) — key-value table, most reliable
+2. **Variant/SKU specs** — per-variant weight or size
+3. **Logistics section** — shipping weight, volume, freight info
+4. **Detail images** — downloaded to `/tmp/1688-logistics/<offer-id>/`, read them to find weight/size text baked into images
+
+## Output
 
 ```json
 {
-  "offerId": "966107271425",
+  "status": "success",
-  "url": "https://detail.1688.com/offer/966107271425.html",
+  "url": "https://detail.1688.com/offer/...",
-  "title": "商品标题",
+  "product": {
-  "weight": {
+    "title": "产品标题",
-    "value": 0.15,
+    "logistics": {
-    "unit": "kg",
+      "weight": { "value": 0.5, "unit": "kg", "source": "attributes" },
-    "source": "商品属性"
+      "dimensions": { "length": 30, "width": 20, "height": 10, "unit": "cm", "source": "attributes" },
+      "grossWeight": null,
+      "netWeight": null,
+      "packageWeight": null,
+      "volume": null,
+      "shippingMethod": null,
+      "shippingCost": null,
+      "origin": null
+    },
+    "variants": [
+      { "name": "颜色: 红色", "weight": null, "dimensions": null }
+    ]
   },
-  "grossWeight": {
+  "detailImages": ["/tmp/1688-logistics/852504650877/img_001.jpg"],
-    "value": 0.2,
+  "rawAttributes": { "重量": "0.5kg", "尺寸": "30*20*10cm" }
-    "unit": "kg",
-    "source": "商品件重尺"
-  },
-  "netWeight": {
-    "value": 0.15,
-    "unit": "kg",
-    "source": "商品属性"
-  },
-  "dimensions": {
-    "length": 10,
-    "width": 8,
-    "height": 1.8,
-    "unit": "cm",
-    "source": "商品属性"
-  },
-  "volume": {
-    "value": 0.000144,
-    "unit": "m³",
-    "source": "商品件重尺"
-  },
-  "packageWeight": {
-    "value": 5.0,
-    "unit": "kg",
-    "source": "包装信息"
-  },
-  "packageDimensions": {
-    "length": 40,
-    "width": 30,
-    "height": 20,
-    "unit": "cm",
-    "source": "包装信息"
-  },
-  "unitsPerPackage": 50,
-  "variants": [
-    {
-      "name": "12支装",
-      "weight": { "value": 0.12, "unit": "kg" },
-      "dimensions": { "length": 9.5, "width": 6, "height": 2.2, "unit": "cm" }
-    }
-  ]
 }
 ```
 
-### Field rules
+`null` = not found in text. Check `detailImages` — the data may be in the images.
-
-- **All weight values normalized to kg** (克÷1000, 斤×0.5)
-- **All dimension values normalized to cm** (mm÷10)
-- **`source`**: where on the page the data was found (商品属性 / 商品件重尺 / 包装信息 / 详情图片)
-- **`variants`**: only include if weight/size differs per SKU. Omit if all variants share the same specs.
-- **Omit fields that are `null`** — do not include fields where no data was found
-- **Do not guess.** Only include values actually visible on the page or in images.
 
 ## Rules
 
 1. If the browser is not running, report the error. Do not try to launch it.
-2. No retries. If it fails, report as-is.
+2. Check all data sources before reporting `null`.
-3. Read ALL screenshots — logistics data can appear anywhere on the page.
+3. Normalize units: 克→kg, 毫米→cm. Keep raw values in `rawAttributes`.
-4. Read detail images too — weight/size is often baked into product photos.
+4. No retries. If it fails, report as-is.
-5. Output ONLY the structured JSON above. No extra commentary.
+5. Trust page content. Do not guess values.

install.sh

@@ -2,26 +2,8 @@
 set -euo pipefail
 cd "$(dirname "$0")"
 
-# Auto-install auth-rt if not found
-if ! command -v auth-rt &>/dev/null && [ ! -x "$HOME/.local/bin/auth-rt" ]; then
-  echo "auth-rt not found, installing..."
-  _FORGEJO="http://192.168.0.108:3030"
-  _OS="$(uname -s | tr '[:upper:]' '[:lower:]')"
-  _ARCH="$(uname -m)"; case "$_ARCH" in x86_64) _ARCH="amd64";; aarch64) _ARCH="arm64";; esac
-  _URL="$_FORGEJO/agent-skills/auth-runtime/releases/download/latest/auth-rt-${_OS}-${_ARCH}"
-  mkdir -p "$HOME/.local/bin"
-  if curl -fsSL "$_URL" -o "$HOME/.local/bin/auth-rt" 2>/dev/null; then
-    chmod +x "$HOME/.local/bin/auth-rt"
-    echo "auth-rt installed (downloaded)"
-  else
-    echo "Download failed, building from source..."
-    _SRC="$HOME/.local/share/auth-runtime"
-    if [ -d "$_SRC/.git" ]; then git -C "$_SRC" pull --ff-only
-    else git clone --depth 1 "$_FORGEJO/agent-skills/auth-runtime.git" "$_SRC"
-    fi
-    bash "$_SRC/install.sh"
-  fi
-fi
-
 bun install
 echo "1688-logistics-scraper installed."
+echo ""
+echo "Prerequisites: Chrome must be running with remote debugging:"
+echo "  /Applications/Google\\ Chrome.app/Contents/MacOS/Google\\ Chrome --remote-debugging-port=9222"

scripts/run.ts

@@ -12,14 +12,17 @@ Commands:
 Examples:
   bun scripts/run.ts scrape 'https://detail.1688.com/offer/852504650877.html'
   bun scripts/run.ts scrape 'https://detail.1688.com/offer/852504650877.html' --dry-run
-  bun scripts/run.ts --port=18801 scrape 'https://detail.1688.com/offer/852504650877.html'
+  bun scripts/run.ts --port=9223 scrape 'https://detail.1688.com/offer/852504650877.html'
+
+Prerequisites:
+  Chrome must be running with --remote-debugging-port=9222
 `);
 }
 
 async function main(): Promise<void> {
   const positionals: string[] = [];
   let dryRun = false;
-  let port = 18800;
+  let port = 9222;
 
   for (const arg of process.argv.slice(2)) {
     if (arg === '--dry-run') {

src/auth-cli.ts

@@ -1,119 +0,0 @@
-/**
- * Thin CLI wrapper for auth-runtime.
- *
- * Copy this file into your skill's src/ directory. It calls the
- * `auth-rt` binary (a standalone Go executable), so the skill has
- * zero npm/runtime dependency on auth-runtime.
- *
- * Prerequisites:
- *   `auth-rt` must be in PATH or at ~/.local/bin/auth-rt
- *   (install.sh handles this automatically)
- *
- * Usage:
- *   import { createSkillClient } from './auth-cli.ts';
- *   const client = createSkillClient();
- *   const res = await client.post('/ecom/tasks/scrape', { url: '...' });
- */
-
-import { spawnSync } from 'child_process';
-import * as path from 'path';
-import * as os from 'os';
-
-const home = process.env.HOME || os.homedir();
-const AUTH_RT_BIN = process.env.AUTH_RT_BIN
-  || (() => {
-    // Check if auth-rt is in PATH
-    const which = spawnSync('which', ['auth-rt'], { encoding: 'utf-8' });
-    if (which.status === 0 && which.stdout.trim()) {
-      return which.stdout.trim();
-    }
-    return path.join(home, '.local', 'bin', 'auth-rt');
-  })();
-
-export interface ApiResponse {
-  status: number;
-  body: string;
-}
-
-export interface SessionResponse {
-  accessToken: string;
-  expiresIn: number;
-  ownerSessionToken?: string;
-  hookUrl?: string;
-  hookToken?: string;
-}
-
-export interface SkillClientOptions {
-  apiBase?: string;
-  dryRun?: boolean;
-}
-
-function runCli(...args: string[]): string {
-  const result = spawnSync(AUTH_RT_BIN, args, {
-    encoding: 'utf-8',
-    timeout: 60_000,
-  });
-
-  if (result.error) {
-    throw new Error(`auth-rt spawn failed: ${result.error.message}`);
-  }
-  if (result.status !== 0) {
-    throw new Error(`auth-rt failed (exit ${result.status}): ${(result.stderr || '').trim()}`);
-  }
-  return (result.stdout || '').trim();
-}
-
-export class SkillClient {
-  private readonly apiBase?: string;
-  private readonly dryRun: boolean;
-
-  constructor(options: SkillClientOptions = {}) {
-    this.apiBase = options.apiBase;
-    this.dryRun = options.dryRun ?? false;
-  }
-
-  async session(): Promise<SessionResponse> {
-    if (this.dryRun) {
-      return { accessToken: '<dry-run-token>', expiresIn: 900 };
-    }
-    return JSON.parse(runCli('session'));
-  }
-
-  async get(urlPath: string): Promise<ApiResponse> {
-    return this.request('GET', urlPath);
-  }
-
-  async post(urlPath: string, body?: unknown): Promise<ApiResponse> {
-    return this.request('POST', urlPath, body);
-  }
-
-  async put(urlPath: string, body?: unknown): Promise<ApiResponse> {
-    return this.request('PUT', urlPath, body);
-  }
-
-  async patch(urlPath: string, body?: unknown): Promise<ApiResponse> {
-    return this.request('PATCH', urlPath, body);
-  }
-
-  async delete(urlPath: string, body?: unknown): Promise<ApiResponse> {
-    return this.request('DELETE', urlPath, body);
-  }
-
-  private async request(method: string, urlPath: string, body?: unknown): Promise<ApiResponse> {
-    if (this.dryRun) {
-      return { status: 200, body: JSON.stringify({ dryRun: true, method, path: urlPath }) };
-    }
-    const args = ['request', method, urlPath];
-    if (body != null) {
-      args.push('--body', JSON.stringify(body));
-    }
-    if (this.apiBase) {
-      args.push('--api-base', this.apiBase);
-    }
-    return JSON.parse(runCli(...args));
-  }
-}
-
-export function createSkillClient(options?: SkillClientOptions): SkillClient {
-  return new SkillClient(options);
-}

src/index.ts

@@ -3,16 +3,50 @@ import * as path from 'path';
 
 export type Command = 'scrape';
 
+export interface LogisticsValue {
+  value: number | null;
+  unit: string | null;
+  source: string;
+}
+
+export interface Dimensions {
+  length: number | null;
+  width: number | null;
+  height: number | null;
+  unit: string | null;
+  source: string;
+}
+
+export interface LogisticsData {
+  weight: LogisticsValue | null;
+  dimensions: Dimensions | null;
+  grossWeight: LogisticsValue | null;
+  netWeight: LogisticsValue | null;
+  packageWeight: LogisticsValue | null;
+  volume: LogisticsValue | null;
+  shippingMethod: string | null;
+  shippingCost: string | null;
+  origin: string | null;
+}
+
+export interface VariantInfo {
+  name: string;
+  weight: LogisticsValue | null;
+  dimensions: Dimensions | null;
+}
+
 export interface ScrapeResult {
   status: 'success' | 'failed';
   url: string;
   command: Command;
   dryRun: boolean;
-  offerId: string;
+  product?: {
-  productPackInfo?: unknown;
+    title: string;
-  windowContext?: unknown;
+    logistics: LogisticsData;
-  screenshots?: string[];
+    variants: VariantInfo[];
+  };
   detailImages?: string[];
+  rawAttributes?: Record<string, string>;
   error?: string;
 }
 
@@ -28,10 +62,9 @@ class CdpSession {
   private ws!: WebSocket;
   private msgId = 0;
   private pending = new Map<number, { resolve: (v: any) => void; reject: (e: Error) => void }>();
-  private eventListeners = new Map<string, Array<(params: any) => void>>();
 
   static async connect(port: number): Promise<CdpSession> {
-    const resp = await fetch(`http://localhost:${port}/json`);
+    const resp = await fetch(`http://127.0.0.1:${port}/json`);
     const targets = (await resp.json()) as Array<{ webSocketDebuggerUrl: string; type: string }>;
     const page = targets.find(t => t.type === 'page');
     if (!page) throw new Error('No Chrome page tab found. Open a tab first.');
@@ -46,20 +79,13 @@ class CdpSession {
       this.ws.onopen = () => resolve();
       this.ws.onerror = (e: any) => reject(new Error(`WebSocket error: ${e.message || e}`));
       this.ws.onmessage = (ev: MessageEvent) => {
-        const msg = JSON.parse(typeof ev.data === 'string' ? ev.data : ev.data.toString());
+        const msg: CdpResult = JSON.parse(typeof ev.data === 'string' ? ev.data : ev.data.toString());
-        // Handle command responses
         if (msg.id != null && this.pending.has(msg.id)) {
           const p = this.pending.get(msg.id)!;
           this.pending.delete(msg.id);
           if (msg.error) p.reject(new Error(msg.error.message));
           else p.resolve(msg.result);
         }
-        // Handle events
-        if (msg.method && this.eventListeners.has(msg.method)) {
-          for (const fn of this.eventListeners.get(msg.method)!) {
-            fn(msg.params);
-          }
-        }
       };
     });
   }
@@ -72,117 +98,145 @@ class CdpSession {
     });
   }
 
-  waitForEvent(event: string, timeoutMs: number = 30000): Promise<any> {
-    return new Promise((resolve, reject) => {
-      const timer = setTimeout(() => {
-        cleanup();
-        reject(new Error(`Timeout waiting for ${event}`));
-      }, timeoutMs);
-      const handler = (params: any) => {
-        cleanup();
-        resolve(params);
-      };
-      const cleanup = () => {
-        clearTimeout(timer);
-        const listeners = this.eventListeners.get(event);
-        if (listeners) {
-          const idx = listeners.indexOf(handler);
-          if (idx >= 0) listeners.splice(idx, 1);
-        }
-      };
-      if (!this.eventListeners.has(event)) this.eventListeners.set(event, []);
-      this.eventListeners.get(event)!.push(handler);
-    });
-  }
-
   async evaluate(expression: string): Promise<any> {
     const res = await this.send('Runtime.evaluate', { expression, returnByValue: true });
     return res?.result?.value;
   }
 
-  async captureScreenshot(format: string = 'png'): Promise<Buffer> {
-    const res = await this.send('Page.captureScreenshot', {
-      format,
-      captureBeyondViewport: false,
-    });
-    return Buffer.from(res.data, 'base64');
-  }
-
   close() {
     try { this.ws.close(); } catch {}
   }
 }
 
-// --- Helpers ---
+// --- Parsers ---
+
+const WEIGHT_KEYS = ['重量', '毛重', '净重', '单件重量', '包装重量', '产品重量', '单品重量', 'weight'];
+const DIMENSION_KEYS = ['尺寸', '规格', '长宽高', '外箱尺寸', '包装尺寸', '产品尺寸', '大小', 'size', 'dimensions'];
+const VOLUME_KEYS = ['体积', '容积', 'volume'];
 
 function extractOfferId(url: string): string {
   return url.match(/offer\/(\d+)/)?.[1] || 'unknown';
 }
 
-async function scrollAndCapture(
+function parseWeight(raw: string): LogisticsValue | null {
-  cdp: CdpSession,
+  const m = raw.match(/([\d.]+)\s*(kg|g|克|千克|公斤|斤)/i);
-  outputDir: string,
+  if (!m) return null;
-): Promise<string[]> {
+  let value = parseFloat(m[1]);
-  fs.mkdirSync(outputDir, { recursive: true });
+  let unit = m[2].toLowerCase();
-  const saved: string[] = [];
+  if (unit === 'g' || unit === '克') { value /= 1000; unit = 'kg'; }
-
+  if (unit === '千克' || unit === '公斤') unit = 'kg';
-  // Get page height
+  if (unit === '斤') { value *= 0.5; unit = 'kg'; }
-  const pageHeight: number = await cdp.evaluate(
+  return { value, unit, source: '' };
-    'Math.max(document.body.scrollHeight, document.documentElement.scrollHeight)'
-  ) || 0;
-  const viewportHeight: number = await cdp.evaluate('window.innerHeight') || 900;
-
-  // Scroll through the page and capture viewport-sized screenshots
-  // Use 80% step to overlap slightly and avoid missing content at boundaries
-  const step = Math.floor(viewportHeight * 0.8);
-  let scrollY = 0;
-  let idx = 1;
-  while (scrollY < pageHeight) {
-    await cdp.evaluate(`window.scrollTo(0, ${scrollY})`);
-    await new Promise(r => setTimeout(r, 800)); // wait for lazy-load render
-
-    const buf = await cdp.captureScreenshot('png');
-    const filePath = path.join(outputDir, `page_${String(idx).padStart(3, '0')}.png`);
-    fs.writeFileSync(filePath, buf);
-    saved.push(filePath);
-
-    scrollY += step;
-    idx++;
-  }
-
-  return saved;
 }
 
-async function downloadDetailImages(
+function parseDimensions(raw: string): Dimensions | null {
-  cdp: CdpSession,
+  const m = raw.match(/([\d.]+)\s*[*xX×]\s*([\d.]+)\s*[*xX×]\s*([\d.]+)\s*(cm|mm|毫米|厘米|m|米)?/i);
-  outputDir: string,
+  if (!m) return null;
-): Promise<string[]> {
+  let [l, w, h] = [parseFloat(m[1]), parseFloat(m[2]), parseFloat(m[3])];
-  // Get all detail image URLs from the page
+  let unit = (m[4] || 'cm').toLowerCase();
-  const imgUrls: string[] = JSON.parse(await cdp.evaluate(`
+  if (unit === 'mm' || unit === '毫米') { l /= 10; w /= 10; h /= 10; unit = 'cm'; }
-    (function() {
+  if (unit === '厘米') unit = 'cm';
-      const imgs = [], seen = new Set();
+  if (unit === 'm' || unit === '米') { l *= 100; w *= 100; h *= 100; unit = 'cm'; }
-      document.querySelectorAll('img').forEach(img => {
+  return { length: l, width: w, height: h, unit, source: '' };
-        const src = img.src || img.dataset.src || img.dataset.lazySrc || '';
+}
-        if (src && !seen.has(src) && (src.startsWith('http') || src.startsWith('//'))) {
-          // Filter for product detail images (skip tiny icons/avatars)
-          if (img.naturalWidth > 200 || img.width > 200 || !img.complete) {
-            seen.add(src);
-            imgs.push(src.startsWith('//') ? 'https:' + src : src);
-          }
-        }
-      });
-      return JSON.stringify(imgs);
-    })()
-  `) || '[]');
 
+function parseVolume(raw: string): LogisticsValue | null {
+  const m = raw.match(/([\d.]+)\s*(m³|cm³|L|ml|升|毫升|立方米|立方厘米)/i);
+  if (!m) return null;
+  return { value: parseFloat(m[1]), unit: m[2], source: '' };
+}
+
+function matchKey(text: string, keys: string[]): boolean {
+  const lower = text.toLowerCase();
+  return keys.some(k => lower.includes(k.toLowerCase()));
+}
+
+// --- Page extraction ---
+
+const JS_EXTRACT_ATTRS = `
+(function() {
+  const attrs = {};
+  const sels = [
+    '.detail-attributes-list .attributes-item',
+    '.obj-leading .obj-content li',
+    '#mod-detail-attributes .attribute-item',
+    '.detail-info table tr',
+    '[class*="attribute"] li',
+    '[class*="param"] li',
+    '.offer-attr-list .offer-attr-item',
+  ];
+  for (const sel of sels) {
+    document.querySelectorAll(sel).forEach(el => {
+      const parts = el.textContent.trim().split(/[:：]/);
+      if (parts.length >= 2) attrs[parts[0].trim()] = parts.slice(1).join(':').trim();
+    });
+  }
+  document.querySelectorAll('table tr, .detail-attributes-list tr').forEach(tr => {
+    const cells = tr.querySelectorAll('td, th');
+    if (cells.length >= 2) attrs[cells[0].textContent.trim()] = cells[1].textContent.trim();
+  });
+  return JSON.stringify(attrs);
+})()`;
+
+const JS_EXTRACT_VARIANTS = `
+(function() {
+  const variants = [];
+  const sels = [
+    '.sku-item-wrapper .sku-item',
+    '[class*="sku"] [class*="item"]',
+    '.obj-sku .obj-content li',
+    '.unit-detail-spec-operator .spec-item',
+  ];
+  for (const sel of sels) {
+    document.querySelectorAll(sel).forEach(el => {
+      const name = el.textContent.trim().replace(/\\s+/g, ' ');
+      if (name && name.length < 200) variants.push({ name, text: el.textContent });
+    });
+  }
+  return JSON.stringify(variants);
+})()`;
+
+const JS_EXTRACT_TITLE = `
+(function() {
+  for (const sel of ['.title-text','.detail-title-text','h1[class*="title"]','.mod-detail-title h1','.d-title']) {
+    const el = document.querySelector(sel);
+    if (el && el.textContent.trim()) return el.textContent.trim();
+  }
+  return document.title || '';
+})()`;
+
+const JS_EXTRACT_IMAGES = `
+(function() {
+  const imgs = [], seen = new Set();
+  const sels = [
+    '#desc-lazyload-container img',
+    '.detail-desc-decorate-richtext img',
+    '[class*="detail-desc"] img',
+    '.mod-detail-description img',
+    '.offer-attr-item img',
+    '.desc-img-loaded img',
+  ];
+  for (const sel of sels) {
+    document.querySelectorAll(sel).forEach(img => {
+      const src = img.src || img.dataset.src || img.dataset.lazySrc || '';
+      if (src && !seen.has(src) && (src.startsWith('http') || src.startsWith('//'))) {
+        seen.add(src);
+        imgs.push(src.startsWith('//') ? 'https:' + src : src);
+      }
+    });
+  }
+  return JSON.stringify(imgs);
+})()`;
+
+async function downloadImages(urls: string[], outputDir: string): Promise<string[]> {
   fs.mkdirSync(outputDir, { recursive: true });
   const saved: string[] = [];
-  for (let i = 0; i < imgUrls.length; i++) {
+  for (let i = 0; i < urls.length; i++) {
     try {
-      const resp = await fetch(imgUrls[i]);
+      const resp = await fetch(urls[i]);
       if (!resp.ok) continue;
       const buf = Buffer.from(await resp.arrayBuffer());
-      const ext = imgUrls[i].match(/\.(jpg|jpeg|png|webp|gif)/i)?.[1] || 'jpg';
+      const ext = urls[i].match(/\.(jpg|jpeg|png|webp|gif)/i)?.[1] || 'jpg';
       const p = path.join(outputDir, `img_${String(i + 1).padStart(3, '0')}.${ext}`);
       fs.writeFileSync(p, buf);
       saved.push(p);
@@ -197,24 +251,30 @@ export async function run(
   command: Command,
   args: string[],
   dryRun: boolean,
-  cdpPort: number = 18800,
+  cdpPort: number = 9222,
 ): Promise<ScrapeResult> {
   if (command !== 'scrape') {
-    return { status: 'failed', url: '', command, dryRun, offerId: '', error: `unknown command: ${command}` };
+    return { status: 'failed', url: '', command, dryRun, error: `unknown command: ${command}` };
   }
 
   const url = args[0];
   if (!url) {
-    return { status: 'failed', url: '', command, dryRun, offerId: '', error: 'scrape requires <url>' };
+    return { status: 'failed', url: '', command, dryRun, error: 'scrape requires <url>' };
   }
 
-  const offerId = extractOfferId(url);
-
   if (dryRun) {
     return {
-      status: 'success', url, command, dryRun, offerId,
+      status: 'success', url, command, dryRun,
-      screenshots: [],
+      product: {
+        title: '<dry-run>',
+        logistics: {
+          weight: null, dimensions: null, grossWeight: null, netWeight: null,
+          packageWeight: null, volume: null, shippingMethod: null, shippingCost: null, origin: null,
+        },
+        variants: [],
+      },
       detailImages: [],
+      rawAttributes: {},
     };
   }
 
@@ -224,76 +284,69 @@ export async function run(
 
     await cdp.send('Page.enable');
     await cdp.send('Runtime.enable');
+    await cdp.send('Page.navigate', { url });
 
-    // Set wide PC viewport to ensure tables fit without horizontal overflow
+    // Wait for load
-    await cdp.send('Emulation.setDeviceMetricsOverride', {
+    await new Promise(r => setTimeout(r, 5000));
-      width: 1920,
+
-      height: 1080,
+    const title: string = await cdp.evaluate(JS_EXTRACT_TITLE) || '';
-      deviceScaleFactor: 2,
+    const rawAttributes: Record<string, string> = JSON.parse(await cdp.evaluate(JS_EXTRACT_ATTRS) || '{}');
-      mobile: false,
+    const rawVariants: Array<{ name: string; text: string }> = JSON.parse(await cdp.evaluate(JS_EXTRACT_VARIANTS) || '[]');
+    const imgUrls: string[] = JSON.parse(await cdp.evaluate(JS_EXTRACT_IMAGES) || '[]');
+
+    const variants: VariantInfo[] = rawVariants.map(v => {
+      const weight = parseWeight(v.text);
+      const dimensions = parseDimensions(v.text);
+      if (weight) weight.source = 'variant';
+      if (dimensions) dimensions.source = 'variant';
+      return { name: v.name, weight, dimensions };
     });
 
-    // Navigate and wait for page load event
+    const logistics: LogisticsData = {
-    const loadPromise = cdp.waitForEvent('Page.loadEventFired', 30000);
+      weight: null, dimensions: null, grossWeight: null, netWeight: null,
-    await cdp.send('Page.navigate', { url });
+      packageWeight: null, volume: null, shippingMethod: null, shippingCost: null, origin: null,
-    await loadPromise;
+    };
 
-    // Wait for networkIdle — poll until no pending requests for 1s
+    for (const [key, val] of Object.entries(rawAttributes)) {
-    await cdp.evaluate(`
+      if (matchKey(key, ['毛重'])) {
-      new Promise(resolve => {
+        logistics.grossWeight = parseWeight(val);
-        let timer;
+        if (logistics.grossWeight) logistics.grossWeight.source = 'attributes';
-        const reset = () => { clearTimeout(timer); timer = setTimeout(resolve, 1000); };
+      } else if (matchKey(key, ['净重'])) {
-        const observer = new PerformanceObserver(() => reset());
+        logistics.netWeight = parseWeight(val);
-        observer.observe({ entryTypes: ['resource'] });
+        if (logistics.netWeight) logistics.netWeight.source = 'attributes';
-        reset();
+      } else if (matchKey(key, ['包装重量'])) {
-      })
+        logistics.packageWeight = parseWeight(val);
-    `);
+        if (logistics.packageWeight) logistics.packageWeight.source = 'attributes';
-
+      } else if (matchKey(key, WEIGHT_KEYS)) {
-    // Extract window.context.result.data
+        logistics.weight = parseWeight(val);
-    let productPackInfo: unknown = null;
+        if (logistics.weight) logistics.weight.source = 'attributes';
-    let windowContext: unknown = null;
+      }
-    const ctx = await cdp.evaluate(`
+      if (matchKey(key, DIMENSION_KEYS)) {
-      (function() {
+        logistics.dimensions = parseDimensions(val);
-        try {
+        if (logistics.dimensions) logistics.dimensions.source = 'attributes';
-          const d = window.context && window.context.result && window.context.result.data;
+      }
-          if (d && d.productPackInfo) {
+      if (matchKey(key, VOLUME_KEYS)) {
-            return JSON.stringify({
+        logistics.volume = parseVolume(val);
-              productPackInfo: d.productPackInfo,
+        if (logistics.volume) logistics.volume.source = 'attributes';
-              productTitle: d.productTitle || null,
+      }
-              productAttributes: d.productAttributes || null,
+      if (matchKey(key, ['产地', '发货地', '所在地'])) {
-              skuSelection: d.skuSelection || null,
+        logistics.origin = val;
-            });
+      }
-          }
-        } catch(e) {}
-        return null;
-      })()
-    `);
-    if (ctx) {
-      const parsed = JSON.parse(ctx);
-      productPackInfo = parsed.productPackInfo;
-      windowContext = parsed;
     }
 
-    const outputDir = path.join('/tmp', '1688-logistics', offerId);
+    const offerId = extractOfferId(url);
-
+    const imgDir = path.join('/tmp', '1688-logistics', offerId);
-    // Capture full-page screenshots (scrolling)
+    const detailImages = await downloadImages(imgUrls, imgDir);
-    const screenshotDir = path.join(outputDir, 'screenshots');
-    const screenshots = await scrollAndCapture(cdp, screenshotDir);
-
-    // Download detail images
-    const imgDir = path.join(outputDir, 'images');
-    const detailImages = await downloadDetailImages(cdp, imgDir);
 
     return {
-      status: 'success', url, command, dryRun, offerId,
+      status: 'success', url, command, dryRun,
-      productPackInfo,
+      product: { title, logistics, variants },
-      windowContext,
-      screenshots,
       detailImages,
+      rawAttributes,
     };
   } catch (error) {
     return {
-      status: 'failed', url, command, dryRun, offerId,
+      status: 'failed', url, command, dryRun,
       error: error instanceof Error ? error.message : String(error),
     };
   } finally {