⏳ This skill is pending AI review.

Scores will appear once the review pipeline completes.

version unknown

weibo-scraper

rrrrrredy@rrrrrredy⭐ 0 stars· last commit 4mo ago· 0 open issues

Weibo public content reader for visible posts that do not require login. Uses web search and Playwright visitor-mode rendering to fetch public user timelines or post text. Triggers: check Weibo, scrape Weibo, Weibo user, read Weibo post, Weibo timeline, fetch Weibo content. Not for: login-required or fans-only content; extracting account cookies; private messages; large-scale batch scraping; non-Weibo platforms.

Choose how to use this skill

You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.

1. Native installer

This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.

Do not guess an installer command or replace an existing version without reviewing the diff.

2. Complete package recommended

Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.

No complete ProSkills package is published for this listing yet.

3. Prompt-only

Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.

Need only the instruction file?

Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.

No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.

—/10

// RATINGS

⭐GitHub Stars
⭐ 0 on GitHubGitHub ↗

New / niche

🟢ProSkills Score
—
📍

Not yet listed on ClawHub or SkillsMP

// README

weibo-scraper

Weibo public content scraper — no login required.

OpenClaw Skill — works with OpenClaw AI agents

What It Does

Fetches public Weibo posts without needing a Weibo account. Uses web search to locate user profiles and post URLs, then opens pages via Playwright with a visitor cookie to extract post text and timestamps. Supports user timeline browsing and individual post content extraction.

Quick Start

openclaw skill install weibo-scraper
# Or:
git clone https://github.com/rrrrrredy/weibo-scraper.git ~/.openclaw/skills/weibo-scraper

Run dependency setup on first use:

bash scripts/setup.sh

Features

  • No login required: visitor cookie mode for public content access
  • User timeline: fetch a user's recent public posts (typically 10-15 visible)
  • Post extraction: read full text content from individual post URLs
  • Web search integration: locate user UIDs and post URLs via site:weibo.com queries
  • Fallback selectors: handles CSS class name changes with alternative extraction methods
  • Hard stop protection: auto-stops after 3 consecutive failures to prevent infinite retries

Usage

Trigger with natural language:

  • "查一下 @某用户 的微博"
  • "看看这个微博用户最近发了什么"
  • "抓取微博内容"
  • "搜微博" / "微博动态"

Project Structure

weibo-scraper/
├── SKILL.md                # Skill definition and workflow
├── scripts/
│   └── setup.sh            # Dependency setup script
├── references/
│   └── weibo-api-notes.md  # API notes and technical details
├── README.md
├── LICENSE
└── .gitignore

Requirements

  • OpenClaw agent runtime
  • Playwright (installed via setup.sh)
  • Web search tool (Bing/Brave/etc.)
  • Upstream proxy for network access (configured in environment)

License

MIT

// HOW IT'S BUILT

// FILES

2 files shipped with this skill. Inline text is the catalog copy. The upstream link is the source repository.

  • SKILL.md5.4 KBupstream ↗
    ---
    name: weibo-scraper
    description: 'Weibo public content reader for visible posts that do not require login. Uses web search and Playwright visitor-mode rendering to fetch public user timelines or post text. Triggers: check Weibo, scrape Weibo, Weibo user, read Weibo post, Weibo timeline, fetch Weibo content. Not for: login-required or fans-only content; extracting account cookies; private messages; large-scale batch scraping; non-Weibo platforms.'
    ---
    
    # weibo-scraper 1.0.0
    
    No login required. Fetches visible public Weibo posts via web search + Playwright visitor-mode rendering.
    
    ---
    
    ## Routing Contract
    
    ### When to use
    
    - User provides a Weibo username, UID, profile URL, or public post URL and asks to read, summarize, or list visible public posts.
    - User asks for a small recent timeline sample or one public post's text/media indicators.
    
    ### When not to use
    
    - User asks for fans-only, login-only, private, deleted, hidden, DM, or account-specific content.
    - User asks to extract cookies, use the user's logged-in account, bypass access controls, or run large-scale scraping.
    - User asks about another platform; route to the matching platform skill.
    
    ### Required inputs
    
    - Username, UID, profile URL, post URL, or enough search terms to locate the public Weibo account.
    - Desired scope: one post, recent N visible posts, or profile overview. Default to recent visible posts only.
    
    ### Output
    
    - Fixed output: source URL/UID, scrape time, fetched count, visible post text, publish time when available, media indicator, and visibility caveats.
    - Do not claim the timeline is complete unless the public page exposes enough chronological evidence.
    
    ### Failure handling
    
    - If search cannot locate the account after three attempts, stop and ask for the profile URL or UID.
    - If Playwright renders an empty page or selectors changed, report the failed URL and selector evidence.
    - If content appears restricted, state the visibility limitation and do not attempt login.
    
    ---
    
    ## ⚠️ Gotchas
    
    ⚠️ **Weibo ajax API returns -100 without login** → `https://m.weibo.cn/api/container/getIndex` etc. always return errno=-100 when not logged in. Do not attempt direct API calls.
    
    ⚠️ **Camoufox may fail to initialize** → Camoufox downloads browser binaries on first run, which can timeout in some network environments. Use standard Playwright instead.
    
    ⚠️ **CSS selector `.detail_wbtext_4CRf9` may change** → Weibo frontend updates class names periodically. When the selector fails, use a fallback: find the timestamp element in page HTML and read surrounding text blocks.
    
    ⚠️ **Search result ordering is not strictly chronological** → Web search results are ranked by relevance, not strict reverse-chronological order. For "absolute latest N posts", open the user's profile page and sort by timestamp manually.
    
    ⚠️ **Unlisted content is inaccessible** → Posts set to "followers only" cannot be fetched. If content appears empty or post count seems low, inform the user about potential visibility restrictions.
    
    ---
    
    ## 🛑 Hard Stop
    
    - Same web search query fails more than **3 times** (empty results or errors) → Stop immediately, tell user: "Cannot locate this user via search. Please verify the Weibo username or provide the profile URL directly."
    - Playwright fails to render content on **3 consecutive** URLs → Stop, inform user the CSS selector may have changed, awaiting manual update.
    
    ---
    
    ## First Use
    
    Run the dependency check before first use:
    ```bash
    bash scripts/setup.sh
    ```
    > This skill depends on Playwright for browser automation and a web search tool.
    
    ---
    
    ## Tool Dependencies
    
    | Tool | Purpose |
    |------|---------|
    | agent-browser (recommended) | Open Weibo pages with visitor cookie to read post content (auto-detects and uses when available) |
    | Playwright | Fallback browser automation when agent-browser is not available |
    | Web search (Bing/Brave/etc.) | Locate Weibo user UID and post URLs via `site:weibo.com` search |
    
    ---
    
    ## Workflow
    
    ### A. Fetch a User's Recent N Posts
    
    **Step 1: Locate user UID**
    ```
    Web search: @username site:weibo.com
    → Extract UID from weibo.com/u/<UID> format URL
    → If post URLs are found directly, skip to Step 3
    ```
    
    **Step 2: Search for user's post list**
    ```
    Web search: site:weibo.com/u/<UID>
    → Collect returned post URLs (typically 5-10)
    ```
    
    **Step 3: Read post content**
    ```
    Playwright: open post URL
    → Read CSS selector: .detail_wbtext_4CRf9
    → If selector fails: find timestamp element in page, read surrounding text blocks
    → Extract: post text + publish time (from page title or timestamp element)
    ```
    
    **Step 4: Format output**
    ```
    Sort by time (descending, based on publish time field)
    Output format:
    ---
    [time] Post content
    (Images/video: [contains media, text shown above])
    ---
    ```
    
    ### B. Get User Profile Overview
    ```
    Playwright: open https://weibo.com/u/<UID> or https://weibo.com/<username>
    → Read post list from page (visitor cookie mode, no login required)
    → Note: without login, visible posts are limited (typically 10-15)
    ```
    
    ---
    
    ## Output Format
    
    ```
    User: @username (UID: XXXXXXXXXX)
    Scraped at: YYYY-MM-DD HH:MM
    ---
    1. [2026-04-06 10:30] Post content...
    2. [2026-04-05 15:20] Post content...
    ...
    Total: N posts fetched, M contain media (text only shown)
    ```
    
    ---
    
    
    > 微博 API 技术笔记见 `references/weibo-api-notes.md`。
    
    ## Changelog
    
    - 1.0.0: Initial release. Web search + Playwright visitor cookie approach for public Weibo content scraping without login.
    
  • README.mdupstream ↗

    Contents are not inlined in the catalog. Open the upstream file.

Plain-text copy, no login: /skill-bundles/rrrrrredy-weibo-scraper.md

// REPO STATS

0 stars
0 open issues
Last commit: 4mo ago