Weibo Scraper - Search, Comments & Brand Monitoring

🔎 What does Weibo Search & Comments Scraper do?

Search public Weibo posts and extract the conversations below them without supplying a Weibo account, phone number, browser cookie, or proxy setting. Enter Chinese or international keywords, brands, products, people, or campaign phrases. The Actor returns normalized posts, public comments, authors, engagement counts, evidence URLs, and optional English translations.

Use relevance search for market research or latest-first search for recurring mention monitoring. You can also pass numeric post IDs to collect a known post and its comments directly.

What you need	What the Actor returns
China brand monitoring	Recent or relevant public mentions and comment threads
Consumer research	Original Chinese text plus best-effort English translation
Campaign analysis	Likes, comments, reposts, authors, and timestamps
Reviewable evidence	Stable post IDs and public evidence URLs
Automation-ready data	One normalized posts-and-comments dataset plus a run summary

🎯 Why use this Weibo scraper?

• Search and comments share one run. The Actor keeps the same managed visitor session from discovery through comment pagination.
• No login fields. Regional routing, visitor-session setup, challenge handling, retries, and throttling are managed internally.
• Post and comment rows are explicit. Use recordType to split the unified dataset without guessing.
• Original text is preserved. Machine-translated English is additive and never replaces the Chinese source.
• Evidence stays attached. Every row includes a public Weibo URL and scrape timestamp.
• Output limits control spend. Cap posts per query and comments per post before starting a run.
• Failures are visible. The OUTPUT record reports request failures instead of silently inventing missing data.

Typical users include China-market researchers, consumer-insight teams, PR and reputation agencies, cross-border brands, investment researchers, academic teams, content strategists, and data engineers.

📦 What Weibo data can the Actor extract?

Post rows can include:

• Search query and numeric post ID
• Public post URL and evidence URL
• Original text and optional English translation
• Publication and scrape timestamps
• Author ID, name, profile URL, verification status, verification reason, and public follower count
• Likes, comments, and reposts
• Image URLs, video URL, hashtags, and public source/location label when exposed

Comment rows can include:

• Parent post ID and URL
• Comment ID and parent comment ID for visible replies
• Original text and optional English translation
• Author ID, name, profile URL, verification status, and public follower count
• Likes, publication time, source/location label, evidence URL, and scrape time

Fields can be null when Weibo does not publish them for a particular row.

🚀 How to scrape Weibo search results and comments

1. Open the Actor and click Try for free.
2. Add one or more search queries. Chinese phrases usually produce the most precise market signal.
3. Choose Relevance for research or Latest for monitoring.
4. Set the maximum posts per query.
5. Keep comment extraction enabled and set a per-post comment limit, or use zero for posts only.
6. Keep English translation enabled if the destination team does not read Chinese.
7. Click Start and open the Posts and comments dataset view.

The prefilled input is intentionally small and returns useful data in one run:

{
    "searchQueries": ["新能源汽车"],
    "sortBy": "relevance",
    "maxPostsPerQuery": 5,
    "includeComments": true,
    "maxCommentsPerPost": 5,
    "translateToEnglish": true
}

⚙️ Input

Input	Description
searchQueries	Keywords, brand names, products, people, hashtags, or campaign phrases
postIds	Optional numeric Weibo post IDs for direct extraction
sortBy	relevance or latest
maxPostsPerQuery	Maximum public posts returned for each query, from 1 to 200
includeComments	Whether to collect public comments for returned or direct posts
maxCommentsPerPost	Comment cap per post, from 0 to 500
translateToEnglish	Adds best-effort English text while retaining original Chinese

API clients may use queries or keywords as aliases for searchQueries, and maxResults as an alias for maxPostsPerQuery.

For direct comment extraction, copy the numeric ID from a mobile Weibo detail URL such as https://m.weibo.cn/detail/5326750785470485.

📤 Output

The default dataset contains both record types. Filter recordType to post or comment, or open the dedicated Comments view.

Example post:

{
    "recordType": "post",
    "query": "新能源汽车",
    "postId": "5326750785470485",
    "postUrl": "https://m.weibo.cn/detail/5326750785470485",
    "createdAt": "2026-08-01T03:20:41.000Z",
    "text": "新能源汽车市场的新变化",
    "textEnglish": "New changes in the new energy vehicle market",
    "authorId": "1639529981",
    "authorName": "公开作者",
    "authorVerified": true,
    "authorFollowersCount": 1779000,
    "likesCount": 1195,
    "commentsCount": 235,
    "repostsCount": 125,
    "hashtags": ["新能源汽车"],
    "scrapedAt": "2026-08-02T10:15:00.000Z",
    "evidenceUrl": "https://m.weibo.cn/detail/5326750785470485"
}

Example comment:

{
    "recordType": "comment",
    "query": "新能源汽车",
    "postId": "5326750785470485",
    "commentId": "5326942993907853",
    "parentCommentId": null,
    "text": "这个变化值得关注",
    "textEnglish": "This change is worth watching",
    "authorName": "公开评论者",
    "likesCount": 4,
    "evidenceUrl": "https://m.weibo.cn/detail/5326750785470485?cid=5326942993907853"
}

The OUTPUT key-value-store record contains query inputs, post and comment counts, total returned records, failed-request count, bounded error details, and generation time. Export the dataset as JSON, CSV, Excel, HTML, or XML.

💵 Pricing

The Actor uses pay-per-event pricing: $0.01 per successfully returned post or comment ($10 per 1,000 records). There is no Actor start fee. Visitor challenges, retries, duplicate rows, failed requests, translations, and the run summary are not charged.

Returned records	Cost
25	$0.25
100	$1.00
1,000	$10.00

Use maxPostsPerQuery, maxCommentsPerPost, and Apify's maximum total charge setting for a strict spending ceiling. The price displayed in Apify Console is authoritative.

🧩 Example Tasks

1. Monitor recent Weibo brand mentions
2. Collect Weibo product feedback
3. Export comments from a Weibo post
4. Research competitor campaign conversations
5. Build a Weibo market research export
6. Track China consumer conversations
7. Watch Weibo reputation-risk keywords
8. Create a posts-only Weibo search feed

Duplicate a Task before adding a schedule, webhook, or destination integration.

🔌 Use the Actor through the API

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('webdata_labs/weibo-search-comments-scraper').call({
    searchQueries: ['新能源汽车', '智能驾驶'],
    sortBy: 'latest',
    maxPostsPerQuery: 20,
    maxCommentsPerPost: 20,
    translateToEnglish: true,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Keep the Apify token in an environment variable or integration secret, never in a prompt or committed file.

🤖 Use with AI agents through MCP

$claude mcp add --transport http apify "https://mcp.apify.com?tools=webdata_labs/weibo-search-comments-scraper"

Claude Desktop, Cursor, or VS Code configuration:

{
    "mcpServers": {
        "apify": {
            "url": "https://mcp.apify.com?tools=webdata_labs/weibo-search-comments-scraper"
        }
    }
}

Example prompt: "Search Weibo for 新能源汽车, collect the latest posts and comments, then summarize recurring product complaints with evidence links."

🔁 Use this Actor in n8n

1. Trigger on a schedule.
2. Start the Actor through the Apify node or an HTTP Request node.
3. Wait for the run to finish and read defaultDatasetId.
4. Fetch dataset items and split them by recordType.
5. Store the snapshot in Sheets, Airtable, Postgres, or a warehouse.
6. Send selected new rows to Slack, email, or an analysis workflow.

Store successive datasets instead of overwriting them when you need mention velocity, persistence, or change detection.

⚠️ Limits and data quality

• Weibo changes public endpoints, experiments, and visitor challenges without notice.
• Public search is a ranked sample, not a complete archive or guaranteed firehose.
• Deleted, private, restricted, moderated, or login-only content is outside scope.
• Visible comment ordering and totals can differ from the counter shown on a post.
• Deep reply threads may be collapsed or unavailable on the public mobile surface.
• Author follower counts can be abbreviated by Weibo and are normalized from values such as 177.9万.
• English text is machine translated and should be reviewed before customer-facing use.
• Location/source labels are platform-provided public strings, not independently verified locations.
• A larger output limit cannot create rows that Weibo does not expose to the current visitor session.

⚖️ Is it legal to scrape Weibo?

The Actor extracts publicly accessible post, comment, and profile metadata. It does not request passwords, private messages, private accounts, or user-supplied account cookies. Public availability does not remove legal, contractual, copyright, privacy, database-right, or platform-term obligations.

Use the data for a legitimate purpose, minimize personal data, set appropriate retention periods, and do not use it for harassment, sensitive profiling, surveillance, or regulated decisions. China's PIPL, GDPR, and other laws may apply depending on the people, jurisdiction, and purpose involved. Obtain legal advice for commercial redistribution or high-risk processing.

❓ FAQ

Do I need a Weibo account or Chinese phone number?

No. The input contains no account credentials or cookie fields. The Actor is limited to public surfaces available to a managed visitor session.

Is this an official Weibo API?

No. It is an independent, unofficial data-extraction Actor and is not affiliated with Weibo or Sina.

Can it sort by newest posts?

Yes. Choose latest. Public search ranking can still include platform-selected results.

Can I scrape comments from one known post?

Yes. Put its numeric ID in postIds, leave comments enabled, and set the desired comment limit.

Why are fewer comments returned than the post counter?

Weibo can collapse replies, moderate rows, require login for deeper pages, or expose only part of a thread publicly.

How do I monitor only new mentions?

Schedule latest-first runs and compare postId and commentId against your own previous datasets. This avoids cross-customer state and makes retention explicit.

How do I report a broken run?

Open the Actor's Issues tab and include the run URL, sanitized input, and expected result. Never post tokens, private cookies, or personal credentials.

🔗 Related Actors

• Xiaohongshu & Douyin Product Trend Radar for Chinese social-commerce discovery.
• 1688 Product Scraper for wholesale products, suppliers, MOQ, and price tiers.
• Google Search Results Scraper for broader demand and competitor validation.
• Meta Ads Library Scraper for paid creative and advertiser monitoring.
• Review Pain Miner API for structured complaints and opportunity themes from review data.
• Website Contact Extractor for public company and supplier contact enrichment.

📝 Changelog

0.1 - 2026-08-02

• Added public Weibo keyword search with relevance and latest modes.
• Added direct post-ID lookup and paginated public comments.
• Added normalized post/comment rows, authors, engagement, media, hashtags, translations, and evidence URLs.
• Added managed China residential visitor sessions, rendered challenge handling, retries, dataset views, summary output, and eight example Task configurations.