# 737 Skill Seekers Db6e7a28 > Skill Seekers - Documentation to AI Skills Automation - Skill: `tools-only/737-skill-seekers-db6e7a28` (Agent Skill, multi-file: 3 files) - Install (CLI): `npx skillmds@latest add tools-only/737-skill-seekers-db6e7a28` - Raw SKILL.md: https://api.skillmd.com/api/skills/tools-only/737-skill-seekers-db6e7a28/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Docs & Writing - Author: tools-only (https://skillmd.com/u/tools-only) - Updated: 2026-09-22 - Page: https://skillmd.com/skills/tools-only/737-skill-seekers-db6e7a28 --- # Skill Seekers - Documentation to AI Skills Automation **Research Date**: January 26, 2026 **Source URL**: **GitHub Repository**: **PyPI Package**: **Version at Research**: v2.7.4 **License**: MIT --- ## Overview Skill Seekers is an open-source Python tool that automatically converts documentation websites, GitHub repositories, and PDF files into production-ready AI skills for Claude, Gemini, OpenAI, and other LLM platforms. **Core Value Proposition**: Transform hours of manual documentation copying into 20-40 minutes of automated skill generation. --- ## Problem Addressed | Problem | How Skill Seekers Solves It | | ------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------- | | Manually copying documentation into AI chat contexts is tedious and error-prone | Automatically extracts from docs, GitHub repos, and PDFs with multi-source analysis | | Large codebases and documentation exceed AI context window limits | Intelligent chunking, categorization, and Three-Stream Analysis (Code, Docs, Insights) | | AI responses lack framework-specific knowledge and best practices | AI-enhanced skills with explanations, examples, and best practices | | Keeping AI knowledge up-to-date with framework changes is time-consuming | Re-run when docs update to keep AI knowledge fresh and accurate | --- ## Key Statistics (as of January 26, 2026) | Metric | Value | | ----------------------- | ------------- | | GitHub Stars | 7,734 | | Forks | 760 | | Contributors | 24 | | Tests Passing | 1,200+ | | Preset Configs | 24 | | LLM Platforms Supported | 4 | | Time per Skill | 20-40 minutes | --- ## Supported Data Sources ### 1. Documentation Websites - **llms.txt Support**: Automatically detects and uses LLM-ready documentation files (10x faster) - **Universal Scraper**: Works with ANY documentation website - **Smart Categorization**: Automatically organizes content by topic - **Code Language Detection**: Recognizes Python, JavaScript, C++, GDScript, etc. ### 2. GitHub Repositories (v2.0.0+) - **Deep Code Analysis**: AST parsing for Python, JavaScript, TypeScript, Java, C++, Go - **API Extraction**: Functions, classes, methods with parameters and types - **Repository Metadata**: README, file tree, language breakdown, stars/forks - **GitHub Issues & PRs**: Fetch open/closed issues with labels and milestones - **CHANGELOG & Releases**: Automatically extract version history - **Conflict Detection**: Compare documented APIs vs actual code implementation ### 3. PDF Files (v1.2.0+) - **Basic PDF Extraction**: Extract text, code, and images from PDF files - **OCR for Scanned PDFs**: Extract text from scanned documents - **Password-Protected PDFs**: Handle encrypted PDFs - **Table Extraction**: Extract complex tables from PDFs - **Parallel Processing**: 3x faster for large PDFs - **Intelligent Caching**: 50% faster on re-runs --- ## Supported Output Platforms | Platform | Format | Auto Upload | AI Enhancement | API Key Required | | -------------------- | ------------------ | ----------- | -------------- | ----------------- | | **Claude AI** | ZIP + YAML | Yes | Yes | ANTHROPIC_API_KEY | | **Google Gemini** | tar.gz | Yes | Yes | GOOGLE_API_KEY | | **OpenAI ChatGPT** | ZIP + Vector Store | Yes | Yes | OPENAI_API_KEY | | **Generic Markdown** | ZIP | No (Manual) | No | None | --- ## Three-Stream GitHub Architecture (v2.6.0+) The tool splits GitHub repository analysis into three complementary streams: ```text ┌─────────────────────────────────────────────────────────────┐ │ GitHub Repository │ └─────────────────────────────────────────────────────────────┘ │ ┌───────────────────┼───────────────────┐ ▼ ▼ ▼ ┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐ │ Stream 1: Code │ │ Stream 2: Docs │ │Stream 3: Insights│ │ │ │ │ │ │ │ - C3.x analysis │ │ - README │ │ - GitHub issues │ │ - Design patterns│ │ - CONTRIBUTING │ │ - Labels │ │ - Test examples │ │ - docs/*.md │ │ - Stars/forks │ │ - Configs │ │ - API docs │ │ - Common problems│ │ - Architecture │ │ │ │ │ └─────────────────┘ └─────────────────┘ └─────────────────┘ ``` **Analysis Depth Options**: - `basic`: 1-2 minutes - Quick overview - `c3x`: 20-60 minutes - Deep analysis with patterns, examples, guides, configs, architecture --- ## C3.x Codebase Analysis System ### C3.1: Design Pattern Extraction Identifies common software design patterns in the codebase. ### C3.2: Test Examples Extracts test cases as usage examples. ### C3.3: AI-Enhanced How-To Guides - Transforms basic guides into professional tutorials - 5 automatic improvements: step descriptions, troubleshooting, prerequisites, next steps, use cases - Dual-mode support: API mode (Claude API) or LOCAL mode (Claude Code CLI) - No API costs with LOCAL mode using Claude Code Max plan ### C3.4: Configuration Pattern Extraction - **9 Config Formats**: JSON, YAML, TOML, ENV, INI, Python, JavaScript, Dockerfile, Docker Compose - **7 Pattern Types**: Database, API, logging, cache, email, auth, server configurations - **Security Analysis**: Finds hardcoded secrets, exposed credentials - **Auto-Documentation**: Generates JSON + Markdown documentation of all configs --- ## Installation Options ```bash # PyPI (Recommended) pip install skill-seekers # uv (Modern, Fast) uv tool install skill-seekers # With Gemini support pip install skill-seekers[gemini] # With OpenAI support pip install skill-seekers[openai] # With all LLM platforms pip install skill-seekers[all-llms] # From Source (Development) git clone https://github.com/yusufkaraaslan/Skill_Seekers.git cd Skill_Seekers pip install -e . ``` --- ## Basic Usage Workflow ```bash # Step 1: Install pip install skill-seekers # Step 2: Scrape documentation skill-seekers scrape --config react # Or from URL directly skill-seekers scrape --url https://react.dev --name react # Step 3: Package and upload skill-seekers package output/react/ skill-seekers upload react.zip # Done! Your skill is ready to use ``` --- ## Multi-Source Unified Scraping (v2.0.0+) Combine documentation + GitHub + PDF in one skill with conflict detection: ```bash # Combine multiple sources skill-seekers scrape \ --url https://react.dev \ --github facebook/react \ --pdf react-patterns.pdf \ --name react-complete ``` **Features**: - Automatically finds discrepancies between docs and code - Rule-based or AI-powered conflict resolution - Side-by-side comparison with warnings - Documentation gap analysis (identifies outdated docs and undocumented features) --- ## Configuration System (v2.7.0+) ### Multi-Token Management ```bash # One-time configuration (5 minutes) skill-seekers config --github # Add multiple GitHub profiles (personal, work, OSS) skill-seekers config # Use specific profile for private repos skill-seekers github --repo mycompany/private-repo --profile work # CI/CD mode (fail fast, no prompts) skill-seekers github --repo owner/repo --non-interactive # View current configuration skill-seekers config --show # Test connections skill-seekers config --test ``` ### Rate Limit Strategies | Strategy | Behavior | Use Case | | ------------------ | ---------------------------------------- | -------------------- | | `prompt` (default) | Ask what to do when rate limited | Interactive use | | `wait` | Automatically wait with countdown timer | Unattended runs | | `switch` | Automatically try next available profile | Multi-account setups | | `fail` | Fail immediately with clear error | CI/CD pipelines | ### Resume Capability ```bash # List resumable jobs skill-seekers resume --list # Resume interrupted job skill-seekers resume github_react_20260117_143022 ``` --- ## Bootstrap Skill - Self-Hosting (v2.7.0+) Generate skill-seekers as a Claude Code skill to use within Claude: ```bash # Generate the skill ./scripts/bootstrap_skill.sh # Install to Claude Code cp -r output/skill-seekers ~/.claude/skills/ # Verify ls ~/.claude/skills/skill-seekers/SKILL.md ``` **Output includes**: - Complete skill documentation - CLI command reference - Quick start examples - Auto-generated API docs - YAML frontmatter validation --- ## Private Config Repositories (v2.2.0+) For team collaboration: - Git-based config sources from private/team git repositories - Multi-source management for GitHub, GitLab, Bitbucket repos - Team collaboration for 3-5 person teams - Enterprise support scaling to 500+ developers - Secure authentication via environment variable tokens - Intelligent caching with automatic pull updates - Offline mode with cached configs --- ## Preset Configurations Available 24 preset configs for popular frameworks: - **Frontend**: React, Vue, Angular, Svelte - **Backend**: Django, FastAPI, Flask, Express - **Game Engines**: Godot (handles 40K+ pages), Unity - **Languages**: Python, JavaScript, TypeScript, Go - **And more...** Browse all at: --- ## MCP Integration Built-in Model Context Protocol support with 27+ tools: ```bash # Setup MCP integration ./setup_mcp.sh ``` Natural language commands: - "Scrape GitHub repo facebook/react" - "Extract config patterns from this codebase" --- ## Relevance to Claude Code Development ### Direct Applications 1. **Skill Generation**: Automate creation of Claude Code skills from any documentation 2. **Documentation-to-Skill Pipeline**: Convert framework docs into AI-consumable formats 3. **Multi-Source Truth**: Combine docs + code + PDFs into unified skill 4. **Conflict Detection**: Identify drift between documentation and implementation ### Patterns Worth Adopting 1. **Three-Stream Analysis**: Separating Code/Docs/Insights for comprehensive coverage 2. **C3.x Depth Levels**: Configurable analysis depth (basic vs deep) 3. **AI Enhancement Modes**: API vs LOCAL modes for cost optimization 4. **Rate Limit Strategies**: Configurable handling for API limits 5. **Resume Capability**: State persistence for interrupted operations ### Integration Opportunities 1. Could integrate with existing skill-creation workflows in this repository 2. Bootstrap capability could generate skills for any Claude Code plugin 3. Configuration patterns could inform plugin settings design 4. Multi-platform export could extend plugin distribution options --- ## Technical Architecture ```text User Input (URL/GitHub/PDF) │ ▼ ┌─────────────────────────────┐ │ Source Detection │ │ (Docs / GitHub / PDF) │ └─────────────────────────────┘ │ ▼ ┌─────────────────────────────┐ │ Content Extraction │ │ - Web scraping │ │ - GitHub API + AST parsing │ │ - PDF text/OCR extraction │ └─────────────────────────────┘ │ ▼ ┌─────────────────────────────┐ │ Analysis & Categorization│ │ - Smart chunking │ │ - Topic detection │ │ - Code language detection │ │ - Conflict detection │ └─────────────────────────────┘ │ ▼ ┌─────────────────────────────┐ │ AI Enhancement │ │ (Optional) │ │ - Add explanations │ │ - Best practices │ │ - Troubleshooting │ └─────────────────────────────┘ │ ▼ ┌─────────────────────────────┐ │ Platform-Specific │ │ Packaging │ │ - Claude ZIP + YAML │ │ - Gemini tar.gz │ │ - OpenAI Vector Store │ │ - Generic Markdown │ └─────────────────────────────┘ │ ▼ ┌─────────────────────────────┐ │ Upload to Platform │ │ (Auto or Manual) │ └─────────────────────────────┘ ``` --- ## References 1. **Official Website**: (accessed 2026-01-26) 2. **GitHub Repository**: (accessed 2026-01-26) 3. **PyPI Package**: (accessed 2026-01-26) 4. **Three-Stream Implementation Summary**: 5. **Claude AI Skills Announcement**: --- ## Freshness Tracking | Field | Value | | ---------------------------- | --------------------- | | Last Verified | 2026-01-26 | | Version at Verification | v2.7.4 | | GitHub Stars at Verification | 7,734 | | Next Review Recommended | 2026-04-26 (3 months) | **Change Detection Indicators**: - Monitor GitHub releases for version changes - Check PyPI for new features - Review changelog for breaking changes - Verify stat changes (stars, forks, contributors)