Skip to content
BranchesTags

Latest commit

 

History

13 Commits

Folders and files

NameName
Last commit message
Last commit date
other-task
other-task
 
 
.gitignore
.gitignore
 
 
README.md
README.md
 
 
launch-browser.js
launch-browser.js
 
 
package-lock.json
package-lock.json
 
 
package.json
package.json
 
 
proxy-utils.js
proxy-utils.js
 
 
run-m4.sh
run-m4.sh
 
 

Repository files navigation

Other-Task Utilities

A collection of web scraping and data analysis utilities for Chinese social media platforms and macro-economic data sources.

📋 Table of Contents

Overview

This repository contains standalone scraping and analysis tools for:

  • Social Media: Twitter/X lists, Weibo groups and user profiles
  • Financial Data: Eastmoney futures market data
  • Macro Economics: Chinese macro-economic indicators from official sources
  • Content Analysis: LLM-powered analysis of collected data

All scrapers use browser automation (Patchright - Playwright fork) with persistent browser profiles to maintain authentication sessions.

Quick Start

Installation

# Install dependencies
npm install

# Create browser profile for authenticated scraping
npm run browser 1

# Run your first scraper
npm run scrape:twitter

Prerequisites

  • Node.js >= 14.0.0
  • Chrome browser installed

Available Tasks

1. Twitter/X Scraper

Scrapes tweets (text + images) from X lists with support for retweets and quote tweets.

Script: other-task/twitter-task/scrape-twitter-list.js

NPM Commands

# Basic usage (default: 5 tweets, profile 1)
npm run scrape:twitter

# With custom arguments
npm run scrape:twitter -- --count 10 --since today
npm run scrape:twitter -- --profile 2 --since 24h
npm run scrape:twitter -- --no-cache  # Force regenerate

Direct Command

node other-task/twitter-task/scrape-twitter-list.js [options]

Parameters

Parameter Type Default Description
--profile <id> number 1 Browser profile ID to use
--count <n> number 5 Number of tweets to scrape
--since <when> string - Time filter (see below)
--url <url> string Built-in Custom X list URL
--output-dir <dir> string twitter-YYYY-MM-DD/ Output directory
--headless flag false Run browser in headless mode
--scroll-delay <ms> number 2000 Delay between scrolls (ms)
--no-cache flag - Force regenerate (ignore cache)
--help flag - Show help message

Time Filters (--since)

  • today - Start of today (midnight)
  • yesterday - Start of yesterday
  • 24h, 12h, 2h, 1h - Relative hours ago
  • 2d, 7d - Relative days ago
  • 30m, 90m - Relative minutes ago
  • YYYY-MM-DD - Specific date (midnight)

Note: Using --since removes the default 5-tweet limit. Add --count N to cap results.

Examples

# Latest 5 tweets (default)
npm run scrape:twitter

# All tweets from today
npm run scrape:twitter -- --since today

# Last 24 hours, max 20 tweets
npm run scrape:twitter -- --since 24h --count 20

# Since specific date
npm run scrape:twitter -- --since 2026-02-09

# Custom list URL with profile 2
npm run scrape:twitter -- --url https://x.com/i/lists/YOUR_LIST_ID --profile 2

# Show browser window (non-headless)
npm run scrape:twitter -- --no-headless

# Force regenerate cached data
npm run scrape:twitter -- --no-cache

User Timeline Scraper

For a single X profile, use:

# Default: https://x.com/aimikoda, down to 2026-02-07 inclusive
npm run scrape:twitter:user

# Resume is enabled by default through crawl-state.json
npm run scrape:twitter:user -- --profile 2

# Start over
npm run scrape:twitter:user -- --fresh

# Custom profile / cutoff
npm run scrape:twitter:user -- --url https://x.com/aimikoda --handle aimikoda --until 2026-02-07

# Optional: also download images
npm run scrape:twitter:user -- --download-images

Output defaults to other-task/twitter-task/twitter-user-aimikoda-2026-02-07/. Each post is saved as blog-0001-<tweet-id>/content.md; progress is recorded in crawl-state.json so interrupted runs skip previously saved post URLs.

Output Structure

twitter-2026-03-03/
  blog-1/
    content.md        # Tweet text with metadata
    image-1.jpg       # High-res images
    image-2.jpg
  blog-2/
    content.md
  ...

Cache Behavior

✅ Cached by date - If twitter-YYYY-MM-DD/ exists, scraping is skipped.
Use --no-cache to force regeneration.

2. Weibo List Scraper

Scrapes posts (text + images) from Weibo group feeds with support for reposts.

Script: other-task/weibo-task/scrape-weibo-list.js

NPM Commands

# Basic usage (default: 5 posts, profile 1)
npm run scrape:weibo

# With custom arguments
npm run scrape:weibo -- --count 10 --since today
npm run scrape:weibo -- --profile 2 --since 2d
npm run scrape:weibo -- --no-cache

Direct Command

node other-task/weibo-task/scrape-weibo-list.js [options]

Parameters

Parameter Type Default Description
--profile <id> number 1 Browser profile ID to use
--count <n> number 5 Number of posts to scrape
--since <when> string - Time filter (same as Twitter)
--url <url> string Built-in Custom Weibo group URL
--output-dir <dir> string weibo-YYYY-MM-DD/ Output directory
--headless flag false Run browser in headless mode
--scroll-delay <ms> number 2000 Delay between scrolls (ms)
--no-cache flag - Force regenerate (ignore cache)
--help flag - Show help message

Examples

# Latest 5 posts (default)
npm run scrape:weibo

# All posts from today
npm run scrape:weibo -- --since today

# Last 2 days, max 20 posts
npm run scrape:weibo -- --since 2d --count 20

# Custom group URL with profile 3
npm run scrape:weibo -- --url https://weibo.com/mygroups/YOUR_GROUP --profile 3

# Force regenerate
npm run scrape:weibo -- --no-cache

Output Structure

weibo-2026-03-03/
  blog-1/
    content.md        # Post text with metadata
    image-1.jpg       # Best quality images
    image-2.jpg
  blog-2/
    content.md
  ...

Cache Behavior

✅ Cached by date - If weibo-YYYY-MM-DD/ exists, scraping is skipped.
Use --no-cache to force regeneration.

3. Weibo User Scraper

Scrapes posts from a specific Weibo user's profile page.

Script: other-task/weibo-task/scrape-weibo-user.js

NPM Commands

# Must provide user URL
npm run scrape:weibo-user -- --url https://weibo.com/u/USER_ID
npm run scrape:weibo-user -- --url https://weibo.com/u/USER_ID --count 20

Direct Command

node other-task/weibo-task/scrape-weibo-user.js <url> [options]
# or
node other-task/weibo-task/scrape-weibo-user.js --url <url> [options]

Parameters

Parameter Type Default Description
--url <url> string Required Weibo user profile URL
<url> string Required Weibo user URL (positional)
--profile <id> number 1 Browser profile ID to use
--count <n> number 50 Max number of posts to scrape
--start <n> number 0 Skip first N posts, start from N+1
--since <when> string - Time filter (same as Twitter)
--output-dir <dir> string weibo-user-{uid}-YYYY-MM-DD/ Output directory
--headless flag false Run browser in headless mode
--scroll-delay <ms> number 2000 Delay between scrolls (ms)
--help flag - Show help message

Examples

# Scrape user's latest 50 posts
npm run scrape:weibo-user -- https://weibo.com/u/1234567890

# Latest 20 posts
npm run scrape:weibo-user -- --url https://weibo.com/u/1234567890 --count 20

# All posts from today
npm run scrape:weibo-user -- https://weibo.com/u/1234567890 --since today

# Skip first 10 posts, scrape next 20
npm run scrape:weibo-user -- https://weibo.com/u/1234567890 --start 10 --count 20

# Use profile 2
npm run scrape:weibo-user -- https://weibo.com/u/1234567890 --profile 2

Output Structure

weibo-user-1234567890-2026-03-03/
  blog-1/
    content.md
    image-1.jpg
  blog-2/
    content.md
  ...

Cache Behavior

❌ No cache - Always fetches fresh data (user posts are frequently updated).

4. Eastmoney Futures Scraper

Scrapes futures market data (news, stock info, forum posts) from Eastmoney.

Script: other-task/eastmoney-task/scrape-eastmoney-futures.js

NPM Commands

# Scrape default symbols (ma, pp, eg, sc)
npm run scrape:eastmoney

# Scrape specific symbols
npm run scrape:eastmoney -- ma pp eg
npm run scrape:eastmoney -- au ag cu  # Gold, silver, copper
npm run scrape:eastmoney -- --no-cache  # Force regenerate

Direct Command

node other-task/eastmoney-task/scrape-eastmoney-futures.js <symbol>... [options]

Parameters

Parameter Type Default Description
<symbol>... string[] ma pp eg sc Futures symbols to scrape
--output-dir <dir> string YYYY-MM-DD_eastmoney_review/ Output directory
--no-headless flag - Show browser window
--no-cache flag - Force regenerate all symbols
--help flag - Show help message

Common Symbols

  • ma - Methanol (甲醇)
  • pp - Polypropylene (聚丙烯)
  • eg - Ethylene Glycol (乙二醇)
  • sc - Crude Oil (原油)
  • ta - PTA
  • au - Gold (黄金)
  • ag - Silver (白银)
  • cu - Copper (铜)

Examples

# Scrape default symbols
npm run scrape:eastmoney

# Scrape specific symbols
npm run scrape:eastmoney -- ma pp

# Scrape metals
npm run scrape:eastmoney -- au ag cu

# Show browser window
npm run scrape:eastmoney -- ma pp --no-headless

# Force regenerate (ignore cache)
npm run scrape:eastmoney -- ma pp eg --no-cache

Output Structure

2026-03-03_eastmoney_review/
  ma/
    news/
      1.md          # Full news articles
      2.md
    stock.md        # Price & market data
    blog.md         # Forum posts (title, author, date, stats)
  pp/
    news/
    stock.md
    blog.md
  ...

Cache Behavior

✅ Cached by date + symbol - Each symbol is cached independently.
Running npm run scrape:eastmoney -- ma pp eg twice will:

  1. First run: Scrape all three (MA, PP, EG)
  2. Second run: Skip all (cached)
  3. Add new symbol: ma pp eg au → Only scrape AU

Use --no-cache to force regenerate all symbols.

5. Macro News Fetcher

Fetches and analyzes macro-economic and industry data from official Chinese sources with optional LLM analysis.

Script: other-task/macro-news-task/fetch-macro-news.js

NPM Commands

# Fetch all sources with LLM analysis
npm run fetch:macro-news

# Fetch macro data only
npm run fetch:macro-news:macro

# Fetch industry data only
npm run fetch:macro-news:industry

# With custom arguments
npm run fetch:macro-news -- --no-llm --concurrency 3
npm run fetch:macro-news -- --source M2_M1_M0 --source CPI_PPI_PMI
npm run fetch:macro-news -- --no-cache  # Force regenerate

Direct Command

node other-task/macro-news-task/fetch-macro-news.js [options]

Parameters

Parameter Type Default Description
--category <type> string all Filter: macro or industry
--source <name> string - Specific data source (repeatable)
--output-dir <path> string ./output Output directory
--headless flag true Run browser in headless mode
--no-headless flag - Show browser window
--no-llm flag - Skip LLM analysis (scrape only)
--no-cache flag - Force regenerate report
--concurrency <n> number 2 Concurrent browsers
--timeout <ms> number 30000 Page load timeout (ms)
--help flag - Show help message

Data Sources

Macro Sources:

  • M2_M1_M0 - Money supply (PBC)
  • CPI_PPI_PMI - Inflation & manufacturing (Stats Bureau)
  • 利率_LPR - Interest rates (PBC)
  • 外汇储备 - Foreign reserves (SAFE)
  • 进出口数据 - Trade data (Customs)
  • 财政收支 - Fiscal data (MOF)
  • 能源产量 - Energy production
  • More...

Industry Sources:

  • 中汽协销量 - Auto sales (CAAM)
  • 港交所交易量 - HKEX trading volume
  • 巨潮资讯网公告 - CNINFO announcements

Examples

# Fetch all sources with LLM
npm run fetch:macro-news

# Macro data only
npm run fetch:macro-news:macro

# Industry data only
npm run fetch:macro-news:industry

# Specific sources
npm run fetch:macro-news -- --source M2_M1_M0 --source CPI_PPI_PMI

# No LLM analysis (faster)
npm run fetch:macro-news -- --no-llm

# Higher concurrency & timeout
npm run fetch:macro-news -- --concurrency 4 --timeout 45000

# Show browser window
npm run fetch:macro-news -- --no-headless

# Force regenerate cached report
npm run fetch:macro-news -- --no-cache

Output

File: output/macro-news-report-YYYY-MM-DD.md

Contents:

  • Table of contents by category (macro/industry)
  • Data source URLs and access status
  • Raw extracted content (collapsed details)
  • AI analysis for each source (if LLM enabled)
  • Statistics summary

LLM Configuration

Create other-task/macro-news-task/.env:

VOLCENGINE_API_KEY=your-volcengine-api-key
VOLCENGINE_MODEL=doubao-pro-32k
LLM_ENDPOINT=https://ark.cn-beijing.volces.com/api/v3
LLM_MAX_TOKENS=2000
LLM_TEMPERATURE=0.7

Both VOLCENGINE_API_KEY and VOLCENGINE_MODEL are required.

Cache Behavior

✅ Cached by date - If macro-news-report-YYYY-MM-DD.md exists, fetching is skipped.
Use --no-cache to force regeneration.

6. Macro Workflow Runner

Runs the macro news fetching as a structured workflow with step-level progress tracking and resume capability.

Script: other-task/macro-news-task/run-workflow.js

NPM Commands

# Run full workflow
npm run workflow:macro-news

# With custom arguments
npm run workflow:macro-news -- --concurrency 3 --no-llm
npm run workflow:macro-news -- --resume  # Resume from saved state
npm run workflow:macro-news -- --no-cache  # Clear state, fresh start

Direct Command

node other-task/macro-news-task/run-workflow.js [options]

Parameters

Parameter Type Default Description
--category <type> string all Filter: macro or industry
--source <name> string - Specific data source (repeatable)
--output-dir <path> string ./output Output directory
--headless flag true Run browser in headless mode
--no-headless flag - Show browser window
--no-llm flag - Skip all LLM analysis steps
--concurrency <n> number 2 Concurrent steps
--timeout <ms> number 30000 Page load timeout (ms)
--retries <n> number 2 Retry count for failed steps
--resume flag - Resume from saved state
--no-cache flag - Clear state file, start fresh
--dry-run flag - Show workflow structure only
--help flag - Show help message

Examples

# Run full workflow
npm run workflow:macro-news

# Resume interrupted workflow
npm run workflow:macro-news -- --resume

# Clear state and restart
npm run workflow:macro-news -- --no-cache

# Show workflow structure without running
npm run workflow:macro-news -- --dry-run

# Custom concurrency and retries
npm run workflow:macro-news -- --concurrency 3 --retries 3

# Skip LLM analysis
npm run workflow:macro-news -- --no-llm

Workflow Steps

Each data source creates a pipeline:

  1. fetch-page - Scrape source URL
  2. analyze - LLM analysis of content
  3. generate-report - Compile final report

State Management

State file: output/workflow-state.json

  • Automatically saves progress after each step
  • --resume continues from last completed step
  • --no-cache clears state for fresh start
  • Each step tracks: status, output, error, retry count

Output

  • State: output/workflow-state.json (progress tracking)
  • Report: output/macro-news-report-YYYY-MM-DD.md (final report)

Cache Behavior

🔄 State-based resume (not traditional cache):

  • Uses workflow-state.json to track step completion
  • Auto-resumes if state exists
  • --resume explicitly continues from state
  • --no-cache clears state for fresh start

Browser Profiles

All scrapers (except Eastmoney) use persistent browser profiles to maintain login sessions.

Creating a Browser Profile

# Launch browser with profile ID
npm run browser 1

# Or directly
node launch-browser.js 1

This opens Chrome with:

  • Persistent storage at browser-profiles/browser-1/
  • Interactive welcome page with login instructions
  • No auto-close (stays open for manual login)

Steps:

  1. Run npm run browser 1
  2. Log in to required platforms:
    • Twitter/X: Visit x.com and log in
    • Weibo: Visit weibo.com and log in
  3. Close browser when done
  4. Your session is saved!

Using Different Profiles

# Create multiple profiles
npm run browser 1  # Profile 1
npm run browser 2  # Profile 2
npm run browser 3  # Profile 3

# Use in scrapers
npm run scrape:twitter -- --profile 1
npm run scrape:twitter -- --profile 2
npm run scrape:weibo -- --profile 3

Use cases:

  • Multiple accounts
  • Different login sessions
  • Proxy rotation
  • Parallel scraping

Cache Behavior

Each task has optimized cache logic:

Task Cache Strategy Auto-invalidate Override
Twitter Date-based Daily --no-cache
Weibo List Date-based Daily --no-cache
Weibo User ❌ No cache Always fresh N/A
Eastmoney Date + Symbol Daily --no-cache
Macro News Date-based Daily --no-cache
Macro Workflow State resume Manual --no-cache

How It Works

Date-based cache:

  • Output directory/file includes date: twitter-2026-03-03/
  • Scraper checks if it exists before running
  • If exists: Fast exit with cache hit message
  • If not exists: Run scraper normally
  • Next day: New date → automatic invalidation

Symbol-based cache (Eastmoney):

  • Each symbol cached independently
  • ma pp cached → Add eg → Only scrapes eg
  • All cached → Fast exit
  • Use --no-cache to regenerate all

No cache (Weibo User):

  • Always fetches fresh data
  • User posts are frequently updated
  • Ensures accuracy

State resume (Workflow):

  • Not a traditional cache
  • Tracks step completion in workflow-state.json
  • --resume continues from last step
  • --no-cache clears state

Cache Examples

# First run: fresh scrape
$ npm run scrape:twitter
🚀 Launching browser...
✅ Scraped 10 tweets

# Second run: cache hit
$ npm run scrape:twitter
📦 Cache hit: twitter-2026-03-03/ already exists
   Contains 10 blog(s)
   Use --no-cache to regenerate

# Force regenerate
$ npm run scrape:twitter -- --no-cache
🗑️  Clearing existing output
🚀 Launching browser...

Proxy Configuration

(Optional) Configure proxies for browser profiles.

Setup

Create proxy-config.json in project root:

[
  {
    "server": "proxy1.example.com:8080",
    "username": "user1",
    "password": "pass1"
  },
  {
    "server": "proxy2.example.com:8080",
    "username": "user2",
    "password": "pass2"
  },
  {
    "server": "proxy3.example.com:8080",
    "username": "user3",
    "password": "pass3"
  }
]

Usage

Profile index maps to proxy array index:

  • Profile 1 → Proxy 1 (array[0])
  • Profile 2 → Proxy 2 (array[1])
  • Profile 3 → Proxy 3 (array[2])
# Create profile 1 with proxy 1
npm run browser 1

# Scrape with profile 2 (uses proxy 2)
npm run scrape:twitter -- --profile 2

Supported Tasks

Proxies work with:

  • ✅ Twitter scraper
  • ✅ Weibo scrapers (list + user)
  • ✅ Browser launcher
  • ❌ Eastmoney (no profile, no proxy)
  • ❌ Macro news (no profile, no proxy)

Troubleshooting

Not Logged In Error

Error: "❌ Not logged in to X/Weibo"

Solution:

  1. Create/refresh profile: npm run browser <profile-id>
  2. Log in to the platform
  3. Close browser
  4. Run scraper again

Browser Profile Not Found

Error: "❌ Browser profile not found"

Solution:

# Create the missing profile
npm run browser 1  # or 2, 3, etc.

Page Not Loading / Timeout

Symptoms: Scraper hangs or times out

Solutions:

  • Check internet connection
  • Increase scroll delay: --scroll-delay 5000
  • Increase timeout (macro only): --timeout 60000
  • Show browser to debug: --no-headless
  • Check if platform changed page structure

Image Download Failures

Symptoms: Some images missing in output

Causes: Network issues, invalid URLs, rate limiting

Solutions:

  • Scraper auto-retries with fallback URLs
  • Check console output for specific errors
  • Reduce concurrency if rate-limited
  • Increase --scroll-delay to give images time to load

LLM Analysis Failures (Macro News)

Error: "LLM call failed" or missing analysis

Solutions:

  1. Verify .env configuration:

    cd other-task/macro-news-task
cat .env  # Should show LLM_API_KEY

  2. Check API quota and rate limits

  3. Skip LLM to test scraping only:

    npm run fetch:macro-news -- --no-llm

  4. Test with single source:

    npm run fetch:macro-news -- --source M2_M1_M0

Cache Issues

Problem: Stale cached data

Solution:

# Force regenerate with --no-cache
npm run scrape:twitter -- --no-cache
npm run fetch:macro-news -- --no-cache
npm run scrape:eastmoney -- ma pp --no-cache

Problem: Want to skip cache permanently

Solution: Delete cached output:

# Twitter
rm -rf other-task/twitter-task/twitter-2026-03-03/

# Weibo
rm -rf other-task/weibo-task/weibo-2026-03-03/

# Eastmoney
rm -rf other-task/eastmoney-task/2026-03-03_eastmoney_review/

# Macro news
rm other-task/macro-news-task/output/macro-news-report-2026-03-03.md

Workflow State Corruption

Problem: Workflow can't resume or shows errors

Solution: Clear state and restart

npm run workflow:macro-news -- --no-cache

Or manually:

rm other-task/macro-news-task/output/workflow-state.json
npm run workflow:macro-news

Debugging Tips

  1. Show browser window: Add --no-headless to see what's happening
  2. Check terminal logs: All scrapers print detailed progress
  3. Test with small counts: Use --count 1 for quick tests
  4. Verify platform access: Open URLs manually in browser
  5. Check rate limits: Reduce --concurrency if hitting limits

NPM Scripts Quick Reference

# Browser Profiles
npm run browser                      # Launch browser profile 1
npm run browser 2                    # Launch browser profile 2

# Twitter
npm run scrape:twitter               # Latest 5 tweets, cache enabled
npm run scrape:twitter -- --count 10 # Latest 10 tweets
npm run scrape:twitter -- --since today  # All tweets from today
npm run scrape:twitter -- --no-cache     # Force regenerate

# Weibo List
npm run scrape:weibo                 # Latest 5 posts, cache enabled
npm run scrape:weibo -- --count 10   # Latest 10 posts
npm run scrape:weibo -- --since today    # All posts from today
npm run scrape:weibo -- --no-cache       # Force regenerate

# Weibo User
npm run scrape:weibo-user -- <url>   # User URL required (no cache)
npm run scrape:weibo-user -- https://weibo.com/u/1234567890

# Eastmoney
npm run scrape:eastmoney             # Default symbols, per-symbol cache
npm run scrape:eastmoney -- ma pp eg # Specific symbols
npm run scrape:eastmoney -- --no-cache   # Force regenerate all

# Macro News
npm run fetch:macro-news             # All sources, cache enabled
npm run fetch:macro-news:macro       # Macro only
npm run fetch:macro-news:industry    # Industry only
npm run fetch:macro-news -- --no-llm     # Skip LLM analysis
npm run fetch:macro-news -- --no-cache   # Force regenerate

# Macro Workflow
npm run workflow:macro-news          # Full workflow with state
npm run workflow:macro-news -- --resume  # Resume from state
npm run workflow:macro-news -- --no-cache    # Clear state, restart
npm run workflow:macro-news -- --dry-run     # Show structure only

Project Structure

article-generator/
├── browser-profiles/          # Persistent browser sessions
│   ├── browser-1/            # Profile 1 data
│   ├── browser-2/            # Profile 2 data
│   └── ...
├── other-task/
│   ├── twitter-task/
│   │   ├── scrape-twitter-list.js
│   │   └── twitter-YYYY-MM-DD/   # Output
│   ├── weibo-task/
│   │   ├── scrape-weibo-list.js
│   │   ├── scrape-weibo-user.js
│   │   ├── weibo-YYYY-MM-DD/     # List output
│   │   └── weibo-user-*-YYYY-MM-DD/  # User output
│   ├── eastmoney-task/
│   │   ├── scrape-eastmoney-futures.js
│   │   └── YYYY-MM-DD_eastmoney_review/  # Output
│   └── macro-news-task/
│       ├── fetch-macro-news.js
│       ├── run-workflow.js
│       ├── data-sources.js
│       ├── llm-client.js
│       ├── workflow-engine.js
│       ├── .env                  # LLM config
│       └── output/
│           ├── macro-news-report-YYYY-MM-DD.md
│           └── workflow-state.json
├── launch-browser.js          # Browser profile launcher
├── proxy-config.json          # Proxy configuration (optional)
├── package.json
└── README.md

License

MIT

Author

yanan.wu

Contributing

Contributions welcome! Please:

  1. Test changes thoroughly
  2. Update documentation
  3. Follow existing code style
  4. Add examples for new features

Support

For issues or questions:

  1. Check Troubleshooting section
  2. Review error messages carefully
  3. Test with --no-headless to see browser behavior
  4. Verify platform website hasn't changed structure

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

4 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

Packages

Contributors

Languages