WeChat Browser Reader

Read WeChat articles via Chrome DevTools when HTTP-based extractors fail.

Prerequisites

• Chrome with remote debugging enabled: google-chrome --no-first-run --remote-debugging-port=9222 --user-data-dir=/tmp/chrome-debug-profile
• OpenClaw browser tools available (navigate_page, evaluate_script, etc.)

Workflow

1. Navigate

navigate_page(url=<article_url>)

2. Handle Captcha (if present)

If the page shows "环境异常" / "去验证":

take_snapshot()  # find the "去验证" element
click(uid=<verify_button_uid>)

If the page shows "轻触查看原文" (non-WeChat container):

evaluate_script(() => document.querySelector('.wx_expand_article_button_wrap')?.click())

3. Wait for Content

WeChat articles use JS encryption. Content may take 3-5 seconds to decrypt after page load.

wait_for(text=["activity-name", "js_content"], timeout=15000)

4. Extract Content

evaluate_script(() => {
  const title = document.getElementById('activity-name')?.innerText || '';
  const author = document.getElementById('js_name')?.innerText || '';
  const content = document.getElementById('js_content')?.innerText || '';
  return { title, author, contentLength: content.length, content };
})

If activity-name is empty but js_content exists, content is loaded — just extract it.

If both are empty after 10+ seconds, try:

evaluate_script(() => new Promise(resolve => {
  setTimeout(() => {
    const el = document.getElementById('js_content');
    resolve({ exists: !!el, htmlLen: el?.innerHTML?.length || 0, text: el?.innerText || '' });
  }, 5000);
}))

5. Return to User

Summarize or present the article content. Key fields:

• title: article title
• author: account name
• content: full article text

Common Issues

Tips

• Always use evaluate_script with setTimeout (3-5s) for reliable content extraction — WeChat's JS decryption is async
• If captcha keeps appearing, the IP may be rate-limited — wait a few minutes
• The approach works because a real Chrome browser executes WeChat's decryption scripts, unlike HTTP-only fetchers