This skill is built with Trellis — an all-in-one AI framework & toolkit for Claude Code, Cursor & iFlow.
Try it out here

---

x-collect

English | 中文

X/Twitter topic intelligence skill for Claude Code. Uses Actionbook browser automation to search x.com directly, scraping real tweets with engagement metrics.

What it does

/x-collect [topic] opens x.com in Chrome, runs 3 search rounds, and extracts real tweet data:

1. Top / Viral Posts — x.com search "Top" tab, most popular tweets
2. Trending / Recent Posts — Top tab with min_faves:50 since:YESTERDAY, hottest posts from the last 24h
3. KOL Posts — filtered by min_faves:100, high-engagement accounts only

Optional bonus rounds:

4. Hook Study — pure text posts (-filter:media -filter:links) with 500+ likes, for studying copy patterns
5. Conversation Starters — high-reply posts (min_replies:30), to find what drives discussion

Output: JSONL + Markdown in ./x-collect-data/ with real handles, tweet text, likes, retweets, replies, views, and a Content Opportunity Summary.

Prerequisites

Dependency	Purpose	Required
Claude Code	Runtime	Yes
Actionbook CLI	Browser automation (actionbook command)	Yes
Actionbook Chrome Extension	Controls user's existing Chrome browser	Yes
Logged-in X/Twitter session	Chrome must have x.com logged in	Yes

Setup Actionbook Extension

Follow the official guide: Actionbook Installation

actionbook extension install           # install extension files
# Then load unpacked extension in Chrome (chrome://extensions)
actionbook extension serve             # start bridge (keep running)

Install

Option A: Claude Plugin Marketplace (Recommended)

claude plugin marketplace add SamCuipogobongo/x-collect
claude plugin install x-collect

Option B: npm global install

npm install -g x-collect-skill

Option C: One-line install

curl -fsSL https://raw.githubusercontent.com/SamCuipogobongo/x-collect/main/install.sh | bash

Option D: Manual

git clone https://github.com/SamCuipogobongo/x-collect.git
mkdir -p ~/.claude/skills/x-collect
cp x-collect/SKILL.md ~/.claude/skills/x-collect/

Usage

# In Claude Code (with Chrome open and x.com logged in):
/x-collect Claude Code       # research "Claude Code" on X
/x-collect vibe coding       # research "vibe coding" on X
/x-collect                    # interactive mode (asks for topic)

Output

Data saved to ./x-collect-data/:

File	Description
intel.jsonl	One JSON object per tweet, deduped by URL
intel.md	Formatted report with engagement data + Content Opportunity Summary

JSONL Schema

{
  "url": "https://x.com/bcherny/status/123",
  "text": "I'm Boris and I created Claude Code...",
  "intel_type": "viral_post",
  "topic": "Claude Code",
  "account": "@bcherny",
  "display_name": "Boris Cherny",
  "angle": "Creator shares vanilla setup with 15+ parallel sessions",
  "format": "thread",
  "likes": 5800,
  "retweets": 1200,
  "replies": 340,
  "views": 6500000,
  "posted_at": "2026-01-15T10:30:00Z",
  "key_takeaway": "Behind-the-scenes from creator + actionable tips = mega engagement",
  "collected": "2026-02-08"
}

Intel Types

Type	Source
viral_post	x.com search "Top" tab
trending_post	Top tab with min_faves:50 since:YESTERDAY
kol_post	x.com search with min_faves:100 filter
hook_study	Pure text posts (-filter:media -filter:links) with 500+ likes
conversation_starter	High-reply posts (min_replies:30)

How it works

The skill uses Actionbook Extension mode to:

1. Control the user's Chrome browser (with existing x.com login)
2. Navigate to x.com search pages with Top tab and various filters
3. Wait for tweets to load
4. Extract tweet data via JavaScript evaluation
5. Deduplicate by tweet URL
6. Output JSONL + Markdown with Content Opportunity Summary

No third-party API needed — reads directly from x.com via Actionbook browser automation.

Community

Questions, feedback, or feature requests? Join our Discord.

License

MIT