{
  "id": "page:docs-reference-repos-yusufkaraaslan-skill-seekers-research",
  "_kind": "Page",
  "_file": "wiki/docs/reference-repos/yusufkaraaslan/skill-seekers/research.md",
  "_cluster": "wiki",
  "attributes": {
    "nodeKind": "Page",
    "sourcePath": "docs/reference-repos/yusufkaraaslan/Skill_Seekers/research.md",
    "sourceKind": "repo-docs",
    "title": "yusufkaraaslan/Skill_Seekers",
    "displayName": "yusufkaraaslan/Skill_Seekers",
    "slug": "docs/reference-repos/yusufkaraaslan/skill-seekers/research",
    "articlePath": "wiki/docs/reference-repos/yusufkaraaslan/Skill_Seekers/research.md",
    "article": "\n# yusufkaraaslan/Skill_Seekers\n\n- **URL**: https://github.com/yusufkaraaslan/Skill_Seekers\n- **Stars**: 12,681\n- **License**: MIT\n- **Last pushed**: 2026-04-12\n- **Description**: Convert documentation websites, GitHub repositories, and PDFs into Claude AI skills with automatic conflict detection\n\n## Overview\n\nSkill Seekers is a \"data layer for AI systems\" -- a universal preprocessing tool that turns documentation sites, GitHub repos, PDFs, videos, notebooks, wikis, and 10+ source types into structured knowledge assets. These assets can then be exported as Claude Skills, Gemini Skills, OpenAI GPTs, LangChain Documents, LlamaIndex TextNodes, Cursor rules, and more.\n\n## Key Features\n\n- 18 input source types (docs sites, GitHub repos, local projects, PDFs, videos, notebooks, wikis, etc.)\n- 20 export targets (12 LLM + 8 RAG/vector): Claude, Gemini, OpenAI, LangChain, LlamaIndex, Haystack, Pinecone, ChromaDB, FAISS, Qdrant, Cursor, etc.\n- AST-based code analysis for structural understanding\n- Smart chunking that preserves code blocks and context\n- Conflict detection for overlapping knowledge\n- MCP server integration\n- 3,194+ tests, 24+ framework presets\n- Claude Code plugin available (`skill-seekers-plugin`)\n- Multi-repo ecosystem: core CLI, website, community configs, GitHub Action, Claude Code plugin, Homebrew tap\n\n## Architecture\n\n- Python CLI (`skill-seekers create <source>`, `skill-seekers package <output> --target <platform>`)\n- Embedding support for vector DB targets\n- Benchmark tooling\n- Workflow system for multi-step processing\n\n## Skills Found\n\n- `skills/skill-seekers/SKILL.md` -- Core skill for the tool itself\n- `distribution/claude-plugin/skills/skill-builder/SKILL.md` -- Skill builder within Claude Code plugin\n- `docs/features/BOOTSTRAP_SKILL.md` -- Bootstrap skill documentation\n\n## Extractable Value for Babysitter\n\n### Plugin Ideas\n1. **Documentation-to-skill converter plugin** -- A babysitter marketplace plugin that wraps Skill Seekers' core capability: point it at documentation (URL, repo, PDF) and automatically generate a SKILL.md with structured knowledge, examples, and patterns. This could be used to bootstrap new babysitter process specializations from existing documentation.\n2. **Knowledge asset indexing plugin** -- Integrate Skill Seekers' multi-source preprocessing into babysitter's process library, enabling automatic knowledge extraction from external documentation into process context.\n\n### Processes (specializations/shared/)\n1. **Documentation-driven skill generation** -- A process that takes a documentation source, extracts structured knowledge, resolves conflicts with existing skills, and produces a validated SKILL.md. The multi-source + conflict-detection + validation pipeline is a reusable methodology.\n\n## Classification\n\n- **Archetype**: Universal documentation-to-skill converter\n- **Primary value**: The concept of automated skill generation from documentation sources with conflict detection and multi-target export. Direct relevance to babysitter's assimilation workflow.\n- **Skip**: Skill management primitives (SDK-covered), the preprocessing engine itself (external tool)\n\n## Library Mapping\n\n| Extractable Process | Library Status | Action | Existing Path | Target Placement |\n|-------------------|----------------|--------|---------------|------------------|\n| Documentation-Driven Skill Generation | NEW | Multi-source knowledge extraction with conflict detection and validation pipeline | - | specializations/shared/documentation-driven-skill-generation.js |\n| Multi-Source Knowledge Extraction | NEW | Universal preprocessing from 18+ source types with AST-based analysis | - | specializations/shared/multi-source-knowledge-extraction.js |\n| Conflict Detection for Knowledge Assets | NEW | Automatic detection and resolution of overlapping knowledge across sources | - | specializations/shared/conflict-detection-knowledge-assets.js |\n| Smart Chunking with Context Preservation | NEW | Intelligent content segmentation that preserves code blocks and contextual relationships | - | specializations/shared/smart-chunking-context-preservation.js |\n\n## Plugin Marketplace Mapping\n\n| Plugin Idea | Marketplace Status | Action | Existing Plugin | Target Placement |\n|-------------|-------------------|--------|-----------------|------------------|\n| Documentation-to-Skill Converter Plugin | NEW | Skill Seekers wrapper for automated skill generation from documentation sources | - | plugins/a5c/marketplace/blueprints/documentation-skill-converter/ |\n| Knowledge Asset Indexing Plugin | NEW | Multi-source preprocessing integration for process library knowledge extraction | - | plugins/a5c/marketplace/blueprints/knowledge-asset-indexing/ |\n",
    "documents": []
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "page:docs-reference-repos",
      "to": "page:docs-reference-repos-yusufkaraaslan-skill-seekers-research",
      "kind": "contains_page"
    }
  ]
}