Turn Docs, Code and PDFs into Claude AI Skills in Minutes (github.com)

Skill Seeker is a new automation tool that converts documentation websites, GitHub repositories and PDFs into production-ready Claude AI skills in minutes. It scrapes any docs site, parses repos with deep AST analysis (Python, JS/TS, Java, C++, Go, etc.), extracts APIs, READMEs, issues/PRs, changelogs and tables, and enhances content with AI to produce organized reference files and a Claude-ready .zip. Features include llms.txt detection, universal scraping, smart categorization, code-language detection, 8 presets (Godot, React, Vue, Django, FastAPI), OCR for scanned PDFs, table extraction, password support, parallel/async scraping, intelligent caching and checkpoint/resume. It also supports MCP/Claude Code integration for natural-language scraping and can run local enhancement with Claude Code Max (no API costs). For the AI/ML community this matters because it automates a time-consuming gap: docs and code drift. Skill Seeker not only packages knowledge quickly (typical 20–40 minute turnaround) but also runs conflict detection (missing-in-code, missing-in-docs, signature and description mismatches) and provides side-by-side reports and rule/AI-powered merges. That yields a single source of truth for teams, faster onboarding for learners, and actionable signals for maintainers and open-source auditors. Technically, its combination of AST-based code analysis, AI-driven example extraction, parallel processing and metadata harvesting (stars, language breakdown) makes it a practical pipeline for keeping LLM tool skills up-to-date and aligned with real implementations.