pip3 install requests beautifulsoup4
Note: Skill_Seekers automatically checks for llms.txt files first, which is 10x faster when available.
Option A: Use a Preset (Easiest)
skill-seekers scrape --config configs/godot.json
Option B: Interactive Mode
skill-seekers scrape --interactive
Option C: Quick Command
skill-seekers scrape --name react --url https://react.dev/
Option D: Unified Multi-Source (NEW - v2.0.0)
# Combine documentation + GitHub code in one skill
skill-seekers unified --config configs/react_unified.json
Detects conflicts between docs and code automatically!
# LOCAL enhancement (no API key, uses Claude Code Max)
skill-seekers enhance output/godot/
This takes 60 seconds and dramatically improves the SKILL.md quality!
skill-seekers package output/godot/
Done! You now have godot.zip ready to use.
# Godot Engine
skill-seekers scrape --config configs/godot.json
# React
skill-seekers scrape --config configs/react.json
# Vue.js
skill-seekers scrape --config configs/vue.json
# Django
skill-seekers scrape --config configs/django.json
# FastAPI
skill-seekers scrape --config configs/fastapi.json
# Unified Multi-Source (NEW!)
skill-seekers unified --config configs/react_unified.json
skill-seekers unified --config configs/django_unified.json
skill-seekers unified --config configs/fastapi_unified.json
skill-seekers unified --config configs/godot_unified.json
If you already scraped once:
skill-seekers scrape --config configs/godot.json
# When prompted:
✓ Found existing data: 245 pages
Use existing data? (y/n): y
# Builds in seconds!
Or use --skip-scrape:
skill-seekers scrape --config configs/godot.json --skip-scrape
# 1. Install (once)
pip3 install requests beautifulsoup4
# 2. Scrape React docs with LOCAL enhancement
skill-seekers scrape --config configs/react.json --enhance-local
# Wait 15-30 minutes (scraping) + 60 seconds (enhancement)
# 3. Package
skill-seekers package output/react/
# 4. Use react.zip in Claude!
Alternative: Enhancement after scraping
# 2a. Scrape only (no enhancement)
skill-seekers scrape --config configs/react.json
# 2b. Enhance later
skill-seekers enhance output/react/
# 3. Package
skill-seekers package output/react/
Edit config file:
{
"max_pages": 20 // Test with just 20 pages
}
# After first scrape, you can rebuild instantly:
skill-seekers scrape --config configs/react.json --skip-scrape
# Copy a preset
cp configs/react.json configs/myframework.json
# Edit it
nano configs/myframework.json
# Use it
skill-seekers scrape --config configs/myframework.json
output/
├── godot_data/ # Raw scraped data (reusable!)
└── godot/ # The skill
├── SKILL.md # With real code examples!
└── references/ # Organized docs
See README.md for:
# Godot
skill-seekers scrape --config configs/godot.json
# Or interactive
skill-seekers scrape --interactive
That's it! 🚀