Invoke

Chinese (Translated)

• "Use x-ai-digest to scrape today's AI hot topics from X"
• "Use x-ai-digest to generate yesterday's X platform AI digest"
• "Use x-ai-digest to check AI-related updates on X from the day before yesterday"
• "Use x-ai-digest to organize recent AI tweets and provide reply suggestions"

English

• "Use x-ai-digest to scrape today's AI hot topics from X"
• "Use x-ai-digest to summarize AI posts from yesterday in English"
• "Generate an X platform AI digest for 2026-01-20 using x-ai-digest"
• "Use x-ai-digest to compile recent AI tweets with reply suggestions"

Japanese (Translated)

• "Use x-ai-digest to summarize today's AI-related posts in Japanese"
• "Use x-ai-digest to obtain yesterday's AI trends on X platform"
• "Use x-ai-digest to organize recent AI-related tweets"
• "Use x-ai-digest to collect AI posts from the day before yesterday and create reply suggestions"

Objectives

• Deliverables: 1 Markdown digest + 1 JSON raw data file + 1 WeChat official account share image
• Each post includes: Time, Author, Content, Original link
• Digest content: Hot topic summary, product releases, selected posts, reply suggestions
• Share image: Card-style design with today's hot topic summary, suitable for sharing on WeChat official accounts/Moments
• Default output: Chinese (English/Japanese content needs to be translated)

Prerequisites

1. dev-browser skill is installed and running
   • Need to use browser extension mode to connect to a browser logged into X
   • Startup command:

     cd skills/dev-browser && npm run start-extension

2. X account is logged in
   • Users need to log into their X account in Chrome browser
   • Ensure access to the "For You" section

Input (Confirm First)

Workflow

Step 1: Connect to Browser and Access X Homepage

// Connect to logged-in browser using dev-browser
const client = await connect();
const page = await client.page("x-home", { viewport: { width: 1920, height: 1080 } });

// Access X homepage
await page.goto("https://x.com/home");
await waitForPageLoad(page);

// Switch to "For You" tab
await page.click('text=For You');
await page.waitForTimeout(3000);

Step 2: Scroll and Scrape Posts

• tweetId

  - Unique post ID

• tweetUrl

  - Original link (

  https://x.com/{username}/status/{tweet_id}

  )

• datetime

  - Publication time in ISO 8601 format

• username

  - Username

• displayName

  - Display name

• content

  - Post content

references/scraping-guide.md

Step 3: Time Filtering

// UTC time range corresponding to Beijing Time YYYY-MM-DD
// Beijing Time 00:00:00 = UTC 16:00:00 of the previous day
// Beijing Time 23:59:59 = UTC 15:59:59 of the same day
const startUTC = new Date(`${date.slice(0, 4)}-${date.slice(5, 7)}-${parseInt(date.slice(8, 10)) - 1}T16:00:00.000Z`);
const endUTC = new Date(`${date}T15:59:59.999Z`);

Step 4: Deduplication and Quantity Control

• Use

  tweetId

  for deduplication
• Stop when 50 posts are accumulated, maximum 100 posts
• If 50 posts are not reached after multiple scrolls, use the actual number
• Stop when no new data is found after 10 consecutive scrolls

Step 5: Filter AI-Related Content

references/ai-keywords.md

Step 6: Generate Digest

assets/digest-template.md

1. Today's hot topics
2. Important product/function releases
3. Industry observations
4. Selected posts and reply suggestions
5. Accounts worth following

• Generate replies in the original language of the post
• Attach Chinese explanations for non-Chinese posts
• Reply content should be valuable, avoid empty compliments

Step 7: Generate WeChat Official Account Share Image

assets/share-card-template.html

1. Generate HTML file: Fill the template with today's hot topics
2. Screenshot to generate PNG: Use playwright to capture the

   .card

   element

import { chromium } from "playwright";

const browser = await chromium.launch({ headless: true });
const page = await browser.newPage({ viewport: { width: 900, height: 1800 } });

await page.goto(`file://${htmlPath}`);
await page.waitForTimeout(2000);

// Capture only the .card element, exclude background
const card = await page.$('.card');
await card.screenshot({
  path: `${outputDir}/ai-digest-${date}.png`,
  type: "png"
});

await browser.close();

• Title and date
• Summaries of 3-4 hot topics today
• Statistical data (number of posts, topics, product releases)
• Data source annotation

Guardrails

• Only scrape publicly visible posts, do not attempt to bind login or bypass restrictions
• Scroll interval >= 1 second to avoid triggering anti-crawling mechanisms
• In case of network errors or page exceptions, downgrade to output the already scraped data
• Confirm the path before outputting files to avoid overwriting existing files
• Keep original links during translation to ensure traceability

Output Examples

File Structure

{output_dir}/
├── x-posts-{date}.json      # Raw data
├── x-posts-{date}.md        # Post list table
├── AI-Digest-{date}.md      # Complete digest
└── ai-digest-{date}.png     # WeChat official account share image

Digest Structure

# Daily AI Digest | {date}

## 1. Today's Hot Topics
### 🔥 Topic 1
### 🆕 Topic 2
...

## 2. Selected Posts and Reply Suggestions
### 1. Author - Topic
**Original**: ...
**Link**: ...
**Suggested Reply (Language)**: ...
**Chinese Explanation**: ... (if non-Chinese)

## 3. Quote of the Day
...

## 4. Accounts Worth Following
...

Resources

• Scraping implementation guide:

  references/scraping-guide.md

• AI keyword list:

  references/ai-keywords.md

• Digest template:

  assets/digest-template.md

• Share card template:

  assets/share-card-template.html

• Share card generation guide:

  references/share-card-guide.md