Common issues and solutions when using Skill Seeker.
Error:
python3: command not found
Solutions:
Check if Python is installed:
which python3
python --version # Try without the 3
Install Python:
brew install python3sudo apt install python3 python3-pipUse python instead of python3:
python cli/doc_scraper.py --help
Error:
ModuleNotFoundError: No module named 'requests'
ModuleNotFoundError: No module named 'bs4'
ModuleNotFoundError: No module named 'mcp'
Solutions:
Install dependencies:
pip3 install requests beautifulsoup4
pip3 install -r mcp/requirements.txt # For MCP
Use --user flag if permission denied:
pip3 install --user requests beautifulsoup4
Check pip is working:
pip3 --version
Error:
Permission denied: '/usr/local/lib/python3.x/...'
Solutions:
Use --user flag:
pip3 install --user requests beautifulsoup4
Use sudo (not recommended):
sudo pip3 install requests beautifulsoup4
Use virtual environment (best practice):
python3 -m venv venv
source venv/bin/activate
pip install requests beautifulsoup4
Error:
FileNotFoundError: [Errno 2] No such file or directory: 'cli/doc_scraper.py'
Solutions:
Check you're in the Skill_Seekers directory:
pwd
# Should show: .../Skill_Seekers
ls
# Should show: README.md, cli/, mcp/, configs/
Change to the correct directory:
cd ~/Projects/Skill_Seekers # Adjust path
Error:
FileNotFoundError: configs/react.json
Solutions:
Check config exists:
ls configs/
# Should show: godot.json, react.json, vue.json, etc.
Use full path:
skill-seekers scrape --config $(pwd)/configs/react.json
Create missing config:
skill-seekers scrape --interactive
Symptoms:
Solutions:
Check configuration file:
cat ~/.config/claude-code/mcp.json
Verify paths are ABSOLUTE (not placeholders):
{
"mcpServers": {
"skill-seeker": {
"args": [
"/Users/yourname/Projects/Skill_Seekers/mcp/server.py"
]
}
}
}
❌ Bad: $REPO_PATH or /path/to/Skill_Seekers
✅ Good: /Users/john/Projects/Skill_Seekers
Test server manually:
cd ~/Projects/Skill_Seekers
python3 mcp/server.py
# Should start without errors (Ctrl+C to stop)
Re-run setup script:
./setup_mcp.sh
# Select "y" for auto-configure
RESTART Claude Code completely:
Problem: Config has $REPO_PATH or /Users/username/ instead of real paths
Solution:
# Get your actual path
cd ~/Projects/Skill_Seekers
pwd
# Copy this path
# Edit config
nano ~/.config/claude-code/mcp.json
# Replace ALL instances of placeholders with your actual path
# Save (Ctrl+O, Enter, Ctrl+X)
# Restart Claude Code
Symptoms:
Solutions:
Check working directory:
{
"cwd": "/FULL/PATH/TO/Skill_Seekers"
}
Verify files exist:
ls cli/doc_scraper.py
ls mcp/server.py
Test CLI tools directly:
skill-seekers scrape --help
Solutions:
Check network connection:
ping google.com
curl -I https://docs.yoursite.com
Use smaller max_pages for testing:
skill-seekers scrape --config configs/test.json --max-pages 5
Increase rate_limit in config:
{
"rate_limit": 1.0 // Increase from 0.5
}
Problem: Pages scraped but content is empty
Solutions:
Check selector in config:
# Test with browser dev tools
# Look for: article, main, div[role="main"], div.content
Verify website is accessible:
curl https://docs.example.com
Try different selectors:
{
"selectors": {
"main_content": "article" // Try: main, div.content, etc.
}
}
Error:
HTTP Error 429: Too Many Requests
Solutions:
Increase rate_limit:
{
"rate_limit": 2.0 // Wait 2 seconds between requests
}
Reduce max_pages:
{
"max_pages": 50 // Scrape fewer pages
}
Try again later:
# Wait an hour and retry
Issue: Can't run ./setup_mcp.sh
Solution:
chmod +x setup_mcp.sh
./setup_mcp.sh
Issue: Homebrew not installed
Solution:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
Issue: pip3 not found
Solution:
sudo apt update
sudo apt install python3-pip
Issue: Permission errors
Solution:
# Use --user flag
pip3 install --user requests beautifulsoup4
Issue: Python not in PATH
Solution:
Issue: Line ending errors
Solution:
dos2unix setup_mcp.sh
./setup_mcp.sh
Use these to check your setup:
# 1. Check Python
python3 --version # Should be 3.10+
# 2. Check dependencies
pip3 list | grep requests
pip3 list | grep beautifulsoup4
pip3 list | grep mcp
# 3. Check files exist
ls cli/doc_scraper.py
ls mcp/server.py
ls configs/
# 4. Check MCP config
cat ~/.config/claude-code/mcp.json
# 5. Test scraper
skill-seekers scrape --help
# 6. Test MCP server
timeout 3 python3 mcp/server.py || echo "Server OK"
# 7. Check git repo
git status
git log --oneline -5
If none of these solutions work:
Check existing issues: https://github.com/yusufkaraaslan/Skill_Seekers/issues
Open a new issue with:
python3 --version)Include this debug info:
# System info
uname -a
python3 --version
pip3 --version
# Skill Seeker info
cd ~/Projects/Skill_Seekers # Your path
pwd
git log --oneline -1
ls -la cli/ mcp/ configs/
# MCP config (if using MCP)
cat ~/.config/claude-code/mcp.json
pwd)python3 --version)pip3 list | grep requests)ls configs/yourconfig.json)ping google.com)$REPO_PATH)Still stuck? Open an issue: https://github.com/yusufkaraaslan/Skill_Seekers/issues/new