Taobao Product Reviews

Taobao — Product Reviews

itemId → paginated customer reviews (reviewer, date, purchased SKU, text, photos)

Language

All process output to user (progress updates, process notifications) follows the user's language.

Objective

Navigate to a Taobao/Tmall product page, load the reviews section, and extract customer review content.

Prerequisites

• Target page is already open in the browser: https://item.taobao.com/item.htm?id={itemId}
• User is logged in to Taobao (user avatar or nickname visible in the page header)

Pre-execution Checks

1. Tool Readiness

If browser-act has been confirmed available in the current session → skip this step.

Invoke browser-act via Skill tool to load usage. If installation or configuration issues arise, follow its guidance to resolve then retry.

2. Login Verification

If login status for Taobao has been confirmed in the current session → skip this step.

Otherwise: open https://www.taobao.com and observe the page header:

• User nickname visible → logged in, continue execution
• Login button visible → not logged in, inform the user that Taobao login is needed first, assist the user in completing the login flow

User refuses or cannot log in → terminate execution.

Capability Components

This Skill's operational boundary = what the user can manually do in their browser. It only reads data already displayed to the user on the page, never bypassing authentication or access controls. JS code is encapsulated in Python files under the scripts/ directory, invoked via eval "$(python scripts/xxx.py {params})". $(...) is bash syntax; it is recommended to use the bash tool for execution.

DOM: product reviews (data extraction)

The reviews section is lazy-loaded below the main product area. Follow these steps to load and extract reviews:

1. navigate "https://item.taobao.com/item.htm?id={itemId}"
2. wait stable
3. Close any popup: look for buttons with text "开心收下", "不了", "关闭" and click to dismiss
4. Scroll to trigger lazy loading of the tabs/reviews section: scroll down --amount 8000
5. wait --selector "[class*='tabTitleItem--']" --state attached --timeout 10000
   • If timeout: scroll down --amount 8000 again and retry wait once more
   • If still no tabs after 2 attempts: take screenshot to confirm page state; the product page may be rendering in a condensed mode — check Known Limitations below
6. eval "$(python scripts/extract-reviews.py '{itemId}')"

Output example:

[
  {
    "username": "一笑奈何",
    "date": "2026-06-03",
    "purchasedSku": "轻巧白|英转中转换器【适用国内电器】适用马来西亚/新加坡等国家",
    "content": "商品非常好，造工很用心！，还会再回购！",
    "photos": [
      "https://gw.alicdn.com/bao/uploaded/i1/O1CN015Cyg4b2FPR2YNq3PD_!!4611686018427383816-0-rate.jpg"
    ],
    "rating": null
  }
]

Notes:

• purchasedSku: the specific variant the reviewer purchased (extracted from "已购：{sku}" prefix in review header)
• content: review text body; may be empty if reviewer submitted only photos
• photos: review photo URLs; empty array if no photos
• rating: star rating; not always visible in current page layout (null is common)
• Reviews shown are the default sort (most recent or most helpful as determined by Taobao)

Error handling: if result count = 0 after scroll attempts, the reviews section may not have loaded in the current browser rendering environment. Try navigating to the product page fresh (navigate again) and repeating the scroll sequence. If still failing, this is a known rendering limitation — see Known Limitations below.

DOM: paginate to next review page

After extracting current page reviews:

1. eval "$(python scripts/next-review-page.py)"
   • Returns {"hasNext": true, "buttonText": "下一页"} if next page exists, or {"hasNext": false} if on last page
2. If hasNext is true: state to find the "下一页" button index → click <index>
3. wait stable
4. Re-run eval "$(python scripts/extract-reviews.py '{itemId}')"

Enum Parameters

[collection failed] Sort/filter options for reviews (e.g., newest, most helpful): these controls exist in the reviews section UI but require the tabs section to be loaded; their URL parameters are not exposed and must be set via UI clicks on the sort tabs within the reviews section.