Why Xeepy?¶
The Problem We Solved¶
In 2023, Twitter (now X) drastically changed their API:
- Basic API: $100/month, severely rate-limited
- Pro API: $5,000/month for reasonable access
- Enterprise: Contact sales (read: very expensive)
Worse, many endpoints were removed entirely. Tweet replies? Gone from the free tier. Full-archive search? Enterprise only. Unfollower detection? Never existed.
The original twitter_reply.py script in this repo used Tweepy's search API—which Twitter broke in 2023. Xeepy was born from the ashes of that broken script.
Our Solution: Browser Automation¶
Instead of fighting API restrictions, we went around them entirely. Xeepy uses Playwright to automate a real browser, giving you the same access as any X user:
# What used to require expensive API access
# now works with simple browser automation
async with Xeepy() as x:
# This "impossible" query now just works
replies = await x.scrape.replies(tweet_url)
Xeepy vs. Alternatives¶
vs. Twitter API (Official)¶
| Aspect | Twitter API | Xeepy |
|---|---|---|
| Monthly Cost | \(100-\)5,000+ | Free |
| Tweet Replies | ❌ Premium only | ✅ Included |
| Full Archive | ❌ Enterprise only | ✅ Included |
| Rate Limits | Strict, per-endpoint | Flexible, human-like |
| Unfollower Detection | ❌ Not available | ✅ Built-in |
| Setup | API keys, OAuth | Just cookies |
vs. Tweepy¶
Tweepy is a great library—but it depends on the Twitter API:
# Tweepy (BROKEN since 2023)
import tweepy
api.search_tweets(q=f"to:{username}") # ❌ No longer works!
# Xeepy (WORKS)
from xeepy import Xeepy
await x.scrape.replies(tweet_url) # ✅ Works perfectly
vs. SNScrape¶
SNScrape was popular but is now unmaintained and broken:
| Aspect | SNScrape | Xeepy |
|---|---|---|
| Status | ❌ Unmaintained | ✅ Actively developed |
| Login Required | No (but limited) | Yes (full access) |
| Anti-Detection | Basic | Advanced |
| Actions | Scraping only | Scraping + Actions |
| AI Features | ❌ No | ✅ Built-in |
vs. Nitter¶
Nitter instances are unreliable and frequently go down:
| Aspect | Nitter | Xeepy |
|---|---|---|
| Reliability | ❌ Instances die often | ✅ Your own browser |
| Authentication | ❌ Can't log in | ✅ Full account access |
| Actions | ❌ Read only | ✅ Full automation |
| Rate Limits | Per-instance | Your control |
Unique Features¶
Xeepy isn't just a scraper—it's a complete automation toolkit:
🎯 Unfollower Detection¶
The #1 requested feature that no API provides:
from xeepy import UnfollowerDetector
detector = UnfollowerDetector(storage, notifier)
report = await detector.detect("yourusername")
print(f"Lost followers: {report.unfollowers}")
print(f"New followers: {report.new_followers}")
print(f"Net change: {report.net_change}")
📊 Advanced Analytics¶
Built-in analytics that would cost $50+/month elsewhere:
from xeepy.analytics import BestTimeAnalyzer, AudienceInsights
# Find YOUR optimal posting times
analyzer = BestTimeAnalyzer()
schedule = await analyzer.analyze("yourusername")
print(schedule.get_schedule_text())
# "Best time to post: Tuesday at 9:00 AM"
# Understand your audience
insights = AudienceInsights()
report = await insights.analyze("yourusername")
print(f"Top locations: {report.locations}")
print(f"Bot percentage: {report.likely_bots_percentage}%")
🤖 AI Integration¶
Native AI support for content and analysis:
from xeepy.ai import ContentGenerator, SentimentAnalyzer
# Generate viral content
generator = ContentGenerator(provider="openai")
thread = await generator.generate_thread(
topic="Python async tips",
style="viral",
num_tweets=10
)
# Analyze sentiment of replies
analyzer = SentimentAnalyzer()
for reply in replies:
sentiment = await analyzer.analyze(reply.text)
if sentiment.label == "negative":
print(f"Hater detected: @{reply.author}")
🔔 Multi-Channel Notifications¶
Get alerts everywhere:
from xeepy.notifications import NotificationManager
manager = NotificationManager()
manager.add_channel("discord", discord_webhook)
manager.add_channel("telegram", telegram_bot)
manager.add_channel("email", email_config)
manager.add_channel("slack", slack_webhook)
# All channels notified instantly
await manager.notify("🚨 Alert!", "Someone important followed you!")
🛡️ Advanced Stealth¶
Undetectable automation:
from xeepy import Xeepy
from xeepy.core import BrowserConfig
config = BrowserConfig(
stealth_mode=True, # Anti-fingerprinting
rotate_user_agent=True, # Random user agents
human_delays=True, # Natural timing
proxy_rotation=True, # IP rotation
)
async with Xeepy(browser_config=config) as x:
# Scrape like a human
...
Who Uses Xeepy?¶
-
Growth Hackers
Automate follow/unfollow, find optimal posting times, track competitor growth.
-
Data Scientists
Scrape large datasets for research, sentiment analysis, network analysis.
-
Marketing Teams
Monitor brand mentions, track campaigns, generate engagement reports.
-
AI Developers
Build AI-powered bots, train models on Twitter data, automate content.
-
Researchers
Academic research, discourse analysis, misinformation tracking.
-
Personal Users
Clean up following list, track unfollowers, optimize posting schedule.
The Bottom Line¶
| Need | Solution |
|---|---|
| Scrape tweet replies | ✅ Xeepy |
| Detect unfollowers | ✅ Xeepy |
| Automate follows | ✅ Xeepy |
| Analyze engagement | ✅ Xeepy |
| Generate AI content | ✅ Xeepy |
| $0 monthly cost | ✅ Xeepy |
Ready to get started?