Files
skill-seekers-reference/docs/QUICK_REFERENCE.md
yusyus 6f1d0a9a45 docs: Comprehensive markdown documentation update for v2.7.0
Documentation Overhaul (7 new files, ~4,750 lines)

Version Consistency Updates:
- Updated all version references to v2.7.0 (ROADMAP.md)
- Standardized test counts to 1200+ tests (README.md, Quality Assurance)
- Updated MCP tool references to 18 tools (CHANGELOG.md)

New Documentation Files:
1. docs/reference/API_REFERENCE.md (750 lines)
   - Complete programmatic usage guide for Python integration
   - All 8 core APIs documented with examples
   - Configuration schema reference and error handling
   - CI/CD integration examples (GitHub Actions, GitLab CI)
   - Performance optimization and batch processing

2. docs/features/BOOTSTRAP_SKILL.md (450 lines)
   - Self-hosting capability documentation (dogfooding)
   - Architecture and workflow explanation (3 components)
   - Troubleshooting and testing guide
   - CI/CD integration examples
   - Advanced usage and customization

3. docs/reference/CODE_QUALITY.md (550 lines)
   - Comprehensive Ruff linting documentation
   - All 21 v2.7.0 fixes explained with examples
   - Testing requirements and coverage standards
   - CI/CD integration (GitHub Actions, pre-commit hooks)
   - Security scanning with Bandit
   - Development workflow best practices

4. docs/guides/TESTING_GUIDE.md (750 lines)
   - Complete testing reference (1200+ tests)
   - Unit, integration, E2E, and MCP testing guides
   - Coverage analysis and improvement strategies
   - Debugging tests and troubleshooting
   - CI/CD matrix testing (2 OS, 4 Python versions)
   - Best practices and common patterns

5. docs/QUICK_REFERENCE.md (300 lines)
   - One-page cheat sheet for quick lookup
   - All CLI commands with examples
   - Common workflows and shortcuts
   - Environment variables and configurations
   - Tips & tricks for power users

6. docs/guides/MIGRATION_GUIDE.md (400 lines)
   - Version upgrade guides (v1.0.0 → v2.7.0)
   - Breaking changes and migration steps
   - Compatibility tables for all versions
   - Rollback instructions
   - Common migration issues and solutions

7. docs/FAQ.md (550 lines)
   - Comprehensive Q&A covering all major topics
   - Installation, usage, platforms, features
   - Troubleshooting shortcuts
   - Platform-specific questions
   - Advanced usage and programmatic integration

Navigation Improvements:
- Added "New in v2.7.0" section to docs/README.md
- Integrated all new docs into navigation structure
- Enhanced "Finding What You Need" section with new entries
- Updated developer quick links (testing, code quality, API)
- Cross-referenced related documentation

Documentation Quality:
- All version references consistent (v2.7.0)
- Test counts standardized (1200+ tests)
- MCP tool counts accurate (18 tools)
- All internal links validated
- Format consistency maintained
- Proper heading hierarchy

Impact:
- 64 markdown files reviewed and validated
- 7 new documentation files created (~4,750 lines)
- 4 files updated (ROADMAP, README, CHANGELOG, docs/README)
- Comprehensive coverage of all v2.7.0 features
- Enhanced developer onboarding experience
- Improved user documentation accessibility

Related Issues:
- Addresses documentation gaps identified in v2.7.0 planning
- Supports code quality improvements (21 ruff fixes)
- Documents bootstrap skill feature
- Provides migration path for users upgrading from older versions

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-18 01:16:22 +03:00

9.1 KiB

Quick Reference - Skill Seekers Cheat Sheet

Version: 2.7.0 | Quick Commands | One-Page Reference


Installation

# Basic installation
pip install skill-seekers

# With all platforms
pip install skill-seekers[all-llms]

# Development mode
pip install -e ".[all-llms,dev]"

CLI Commands

Documentation Scraping

# Scrape with preset config
skill-seekers scrape --config react

# Scrape custom site
skill-seekers scrape --base-url https://docs.example.com --name my-framework

# Rebuild without re-scraping
skill-seekers scrape --config react --skip-scrape

# Async scraping (2-3x faster)
skill-seekers scrape --config react --async

GitHub Repository Analysis

# Basic analysis
skill-seekers github https://github.com/facebook/react

# Deep C3.x analysis (patterns, tests, guides)
skill-seekers github https://github.com/vercel/next.js --analysis-depth c3x

# With GitHub token (higher rate limits)
GITHUB_TOKEN=ghp_... skill-seekers github https://github.com/org/repo

PDF Extraction

# Extract from PDF
skill-seekers pdf manual.pdf --name product-manual

# With OCR (scanned PDFs)
skill-seekers pdf scanned.pdf --enable-ocr

# Large PDF (chunked processing)
skill-seekers pdf large.pdf --chunk-size 50

Multi-Source Scraping

# Unified scraping (docs + GitHub + PDF)
skill-seekers unified --config configs/unified/react-unified.json

# Merge separate sources
skill-seekers merge-sources \
  --docs output/react-docs \
  --github output/react-github \
  --output output/react-complete

AI Enhancement

# API mode (fast, costs ~$0.15-0.30)
export ANTHROPIC_API_KEY=sk-ant-...
skill-seekers enhance output/react/

# LOCAL mode (free, uses Claude Code Max)
skill-seekers enhance output/react/ --mode LOCAL

# Background enhancement
skill-seekers enhance output/react/ --background

# Monitor background enhancement
skill-seekers enhance-status output/react/ --watch

Packaging & Upload

# Package for Claude AI
skill-seekers package output/react/ --target claude

# Package for all platforms
for platform in claude gemini openai markdown; do
  skill-seekers package output/react/ --target $platform
done

# Upload to Claude AI
export ANTHROPIC_API_KEY=sk-ant-...
skill-seekers upload output/react-claude.zip --target claude

# Upload to Google Gemini
export GOOGLE_API_KEY=AIza...
skill-seekers upload output/react-gemini.tar.gz --target gemini

Complete Workflow

# One command: fetch → scrape → enhance → package → upload
export ANTHROPIC_API_KEY=sk-ant-...
skill-seekers install react --target claude --enhance --upload

# Multi-platform install
skill-seekers install react --target claude,gemini,openai --enhance --upload

# Without enhancement or upload
skill-seekers install vue --target markdown

Common Workflows

Workflow 1: Quick Skill from Docs

# 1. Scrape documentation
skill-seekers scrape --config react

# 2. Package for Claude
skill-seekers package output/react/ --target claude

# 3. Upload to Claude
export ANTHROPIC_API_KEY=sk-ant-...
skill-seekers upload output/react-claude.zip --target claude

Workflow 2: GitHub Repo to Skill

# 1. Analyze repository with C3.x features
skill-seekers github https://github.com/facebook/react --analysis-depth c3x

# 2. Package for multiple platforms
skill-seekers package output/react/ --target claude,gemini,openai

Workflow 3: Complete Multi-Source Skill

# 1. Create unified config (configs/unified/my-framework.json)
{
  "name": "my-framework",
  "sources": {
    "documentation": {"type": "docs", "base_url": "https://docs..."},
    "github": {"type": "github", "repo_url": "https://github..."},
    "pdf": {"type": "pdf", "pdf_path": "manual.pdf"}
  }
}

# 2. Run unified scraping
skill-seekers unified --config configs/unified/my-framework.json

# 3. Enhance with AI
skill-seekers enhance output/my-framework/

# 4. Package and upload
skill-seekers package output/my-framework/ --target claude
skill-seekers upload output/my-framework-claude.zip --target claude

MCP Server

Starting MCP Server

# stdio mode (Claude Code, VS Code + Cline)
skill-seekers-mcp

# HTTP mode (Cursor, Windsurf, IntelliJ)
skill-seekers-mcp --transport http --port 8765

MCP Tools (18 total)

Core Tools:

  1. list_configs - List preset configurations
  2. generate_config - Generate config from docs URL
  3. validate_config - Validate config structure
  4. estimate_pages - Estimate page count
  5. scrape_docs - Scrape documentation
  6. package_skill - Package to .zip
  7. upload_skill - Upload to platform
  8. enhance_skill - AI enhancement
  9. install_skill - Complete workflow

Extended Tools: 10. scrape_github - GitHub analysis 11. scrape_pdf - PDF extraction 12. unified_scrape - Multi-source scraping 13. merge_sources - Merge docs + code 14. detect_conflicts - Find discrepancies 15. split_config - Split large configs 16. generate_router - Generate router skills 17. add_config_source - Register git repos 18. fetch_config - Fetch configs from git


Environment Variables

# Claude AI (default platform)
export ANTHROPIC_API_KEY=sk-ant-...

# Google Gemini
export GOOGLE_API_KEY=AIza...

# OpenAI ChatGPT
export OPENAI_API_KEY=sk-...

# GitHub (higher rate limits)
export GITHUB_TOKEN=ghp_...

Testing

# Run all tests (1200+)
pytest tests/ -v

# Run with coverage
pytest tests/ --cov=src/skill_seekers --cov-report=html

# Fast tests only (skip slow tests)
pytest tests/ -m "not slow"

# Specific test category
pytest tests/test_mcp*.py -v             # MCP tests
pytest tests/test_*_integration.py -v    # Integration tests
pytest tests/test_*_e2e.py -v            # E2E tests

Code Quality

# Linting with Ruff
ruff check .                 # Check for issues
ruff check --fix .           # Auto-fix issues
ruff format .                # Format code

# Run before commit
ruff check . && ruff format --check . && pytest tests/ -v

Preset Configurations (24)

Web Frameworks:

  • react, vue, angular, svelte, nextjs

Python:

  • django, flask, fastapi, sqlalchemy, pytest

Game Development:

  • godot, pygame, unity

Tools & Libraries:

  • docker, kubernetes, terraform, ansible

Unified (Docs + GitHub):

  • react-unified, vue-unified, nextjs-unified, etc.

List all configs:

skill-seekers list-configs

Tips & Tricks

Speed Up Scraping

# Use async mode (2-3x faster)
skill-seekers scrape --config react --async

# Rebuild without re-scraping
skill-seekers scrape --config react --skip-scrape

Save API Costs

# Use LOCAL mode for free AI enhancement
skill-seekers enhance output/react/ --mode LOCAL

# Or skip enhancement entirely
skill-seekers install react --target claude --no-enhance

Large Documentation

# Generate router skill (>500 pages)
skill-seekers generate-router output/large-docs/

# Split configuration
skill-seekers split-config configs/large.json --output configs/split/

Debugging

# Verbose output
skill-seekers scrape --config react --verbose

# Dry run (no actual scraping)
skill-seekers scrape --config react --dry-run

# Show config without scraping
skill-seekers validate-config configs/react.json

Batch Processing

# Process multiple configs
for config in react vue angular svelte; do
  skill-seekers install $config --target claude
done

# Parallel processing
skill-seekers install react --target claude &
skill-seekers install vue --target claude &
wait

File Locations

Configurations:

  • Preset configs: skill-seekers-configs/official/*.json
  • Custom configs: configs/*.json

Output:

  • Scraped data: output/{name}_data/
  • Built skills: output/{name}/
  • Packages: output/{name}-{platform}.{zip|tar.gz}

MCP:

  • Server: src/skill_seekers/mcp/server.py
  • Tools: src/skill_seekers/mcp/tools/*.py

Tests:

  • All tests: tests/test_*.py
  • Fixtures: tests/fixtures/

Error Messages

Error Meaning Solution
NetworkError Connection failed Check URL, internet connection
InvalidConfigError Bad config Validate with validate-config
RateLimitError Too many requests Increase rate_limit in config
ScrapingError Scraping failed Check selectors, URL patterns
APIError Platform API failed Check API key, quota

Getting Help

# Command help
skill-seekers --help
skill-seekers scrape --help
skill-seekers install --help

# Version info
skill-seekers --version

# Check configuration
skill-seekers validate-config configs/my-config.json

Documentation:

Links:


Version: 2.7.0 | Test Count: 1200+ | Platforms: Claude, Gemini, OpenAI, Markdown