---

name: weread description: | Use this skill when the user wants to search books on WeRead (微信读书), look up book details, check book availability, or find books by title/author/keyword on weread.qq.com. Also trigger when the user mentions "微信读书", "weread", or wants to check if a book is available on WeRead.

WeRead 微信读书搜索

Search books on WeRead (微信读书) using the public API.

Instructions

When the user invokes /weread <query>, search WeRead and present the results.

Step 1: Parse the query

The argument after /weread is the search keyword. Examples:

• /weread 烟霞洞 → search for "烟霞洞"
• /weread 三体 刘慈欣 → search for "三体 刘慈欣"
• /weread bookId:31131759 → fetch details for a specific book

Step 2: Call the API

Use the Bash tool to call the WeRead search API:

curl -s 'https://weread.qq.com/web/search/global?keyword=ENCODED_KEYWORD' \
  -H 'User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36' \
  -H 'Accept: application/json' \
  -H 'Referer: https://weread.qq.com/'

Replace ENCODED_KEYWORD with the URL-encoded search keyword. Use python3 -c "import urllib.parse; print(urllib.parse.quote('关键词'))" to encode if needed.

Step 3: Parse and display results

Parse the JSON response with a Python one-liner or jq. The response structure:

{
  "books": [
    {
      "bookInfo": {
        "bookId": "...",
        "title": "...",
        "author": "...",
        "cover": "https://cdn.weread.qq.com/...",
        "intro": "...",
        "publisher": "...",
        "price": 0,          // in cents (分)
        "bookStatus": 1,     // 1=available, 5=pending
        "soldout": 0,        // 0=available, 1=removed
        "payType": 4097,
        "format": "epub",
        "newRating": 797,    // divide by 10 for percentage (79.7%)
        "newRatingCount": 3930
      },
      "bookContentInfo": {          // present when keyword found in book content
        "chapterTitle": "...",
        "abstract": "...",          // text snippet with keyword context
        "keyword": ["烟霞洞"]
      },
      "readingCount": 15043,        // daily readers
      "subscribeCount": 1
    }
  ],
  "totalCount": 847,
  "hasMore": true
}

Step 4: Format the output

Present results as a markdown table with these columns:

| # | Title | Author | Rating | Daily Readers | Status | Price | Link | |---|-------|--------|--------|---------------|--------|-------|------|

Rules for formatting:

• Rating: newRating / 10% (e.g., 797 → 79.7%). Show "—" if not available.
• Daily Readers: readingCount. Show "—" if 0.
• Status:
  • bookStatus=1 + soldout=0 → "可读"
  • bookStatus=5 or soldout=1 → "待上架"
• Price: price / 100 元. If 0, show "免费".
• Link: https://weread.qq.com/web/bookDetail/{bookId}
• If bookContentInfo exists, show the matching chapter and text snippet below the table for that book.

Additional capabilities

Book detail lookup

If the query starts with bookId:, fetch book details:

curl -s 'https://weread.qq.com/web/book/info?bookId=BOOK_ID' \
  -H 'User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36' \
  -H 'Referer: https://weread.qq.com/'

Book chapter list

curl -s 'https://weread.qq.com/web/book/chapter/e_0?bookId=BOOK_ID' \
  -H 'User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36' \
  -H 'Referer: https://weread.qq.com/'

Multiple searches

If the user provides multiple keywords separated by |, run each as a separate search and combine results.

API Reference

| Endpoint | Method | Params | Description | |----------|--------|--------|-------------| | /web/search/global | GET | keyword | Search books by keyword | | /web/book/info | GET | bookId | Get book details | | /web/book/chapter/e_0 | GET | bookId | Get chapter list | | /web/book/publicinfos | GET | bookIds (comma-sep) | Batch book info |

All endpoints:

• Base URL: https://weread.qq.com
• No auth required for search
• Set Referer: https://weread.qq.com/ header
• Rate limit: add 1-2s delay between batch requests