X Monitor

x-monitor

Monitor X/Twitter tweet replies and get AI-analyzed answers for technical questions.

An [OpenClaw](https://github.com/openclaw/openclaw) skill.

What It Does

1. Monitor replies — Watch a tweet's comment section for new replies
2. Detect questions — Identify technical questions in replies
3. Draft answers — AI analyzes the question and drafts an answer for you to review

You stay in control — the skill only reads and analyzes, never posts.

Quick Start

# Check replies on a tweet
python3 scripts/monitor.py --url "https://x.com/user/status/123456"

# Watch for new replies (outputs new ones since last check)
python3 scripts/monitor.py --url "https://x.com/user/status/123456" --watch

Requirements

• Python 3.7+ (stdlib only)
• Camofox running on port 9377
• A working Nitter instance (default: nitter.net)

How It Works

1. Uses Camofox browser to render nitter.net (JS-rendered page)
2. Extracts reply author, text, and engagement from the rendered page
3. Compares with previous check to find new replies
4. Optionally identifies technical questions for AI analysis

Output Format

{
  "tweet_url": "https://x.com/user/status/123",
  "checked_at": "2026-02-14T22:00:00",
  "total_replies": 5,
  "new_replies": [
    {
      "author": "@username",
      "text": "How do I fix this error?",
      "likes": 2,
      "media": ["https://pbs.twimg.com/media/XXX.jpg"],
      "is_question": true
    }
  ]
}

File Structure

x-monitor/
├── SKILL.md                    # Agent skill file
├── README.md                   # This file
├── scripts/
│   ├── monitor.py              # Main monitor script (replies + watch mode)
│   └── version_check.py        # Background version checker (silent on startup)
├── data/                       # State files (last-seen reply timestamps)
└── VERSION

Companion Skills

• x-tweet-fetcher