{"id":"skill-builder","name":"skill-builder","summary":"スキルシーカーを使ってソースタイプを自動的に検出し、AIスキルを構築する。ドキュメント、リポジトリ、PDF、動画、その他の知識ソースからスキルを作りたい場合に利用してください。","body":"# Skill Builder\n\nThis skill uses the Skill Seekers MCP server, which provides 40 tools for converting knowledge sources into AI-ready skills. If the MCP tools are not available, use the CLI fallback at the bottom of this file instead — do not stop.\n\n## Prerequisites\n\nThe MCP tools below only work when the Skill Seekers MCP server is connected:\n\n1. Install the package: `pip install \"skill-seekers[mcp]\"`\n2. Connect the server:\n   - Installed as the Skill Seekers plugin? Nothing to do — the plugin's bundled `.mcp.json` starts the server automatically (it still needs step 1).\n   - Installed standalone (e.g. copied into `~/.claude/skills/`)? Register the server once: `claude mcp add skill-seekers -- python -m skill_seekers.mcp.server_fastmcp`\n\nIf tools like `scrape_docs` or `package_skill` are not in your tool list, the server is not connected. Tell the user about the two steps above, and use the CLI fallback in the meantime.\n\n## When to Use This Skill\n\nUse this skill when the user:\n- Wants to create an AI skill from a documentation site, GitHub repo, PDF, video, or other source\n- Needs to convert documentation into a format suitable for LLM consumption\n- Wants to update or sync existing skills with their source documentation\n- Needs to export skills to vector databases (Weaviate, Chroma, FAISS, Qdrant)\n- Asks about scraping, converting, or packaging documentation for AI\n\n## Source Type Detection\n\nAutomatically detect the source type from user input:\n\n| Input Pattern | Source Type | Tool to Use |\n|---------------|-------------|-------------|\n| `https://...` (not GitHub/YouTube) | Documentation | `scrape_docs` |\n| `owner/repo` or `github.com/...` | GitHub | `scrape_github` |\n| `*.pdf` | PDF | `scrape_pdf` |\n| YouTube/Vimeo URL or video file | Video | `scrape_video` |\n| Local directory path | Codebase | `scrape_codebase` |\n| `*.ipynb`, `*.html`, `*.yaml` (OpenAPI), `*.adoc`, `*.pptx`, `*.rss`, `*.1`-`.8` | Various | `scrape_generic` |\n| JSON config file | Unified | Use config with `scrape_docs` |\n\n## Recommended Workflow\n\n1. **Detect source type** from the user's input\n2. **Generate or fetch config** using `generate_config` or `fetch_config` if needed\n3. **Estimate scope** with `estimate_pages` for documentation sites\n4. **Scrape the source** using the appropriate scraping tool\n5. **Enhance** with `enhance_skill` if the user wants AI-powered improvements\n6. **Package** with `package_skill` for the target platform\n7. **Export to vector DB** if requested using `export_to_*` tools\n\n## Available MCP Tools\n\n### Config Management\n- `generate_config` — Generate a scraping config from a URL\n- `list_configs` — List available preset configs\n- `validate_config` — Validate a config file\n\n### Scraping (use based on source type)\n- `scrape_docs` — Documentation sites\n- `scrape_github` — GitHub repositories\n- `scrape_pdf` — PDF files\n- `scrape_video` — Video transcripts\n- `scrape_codebase` — Local code analysis\n- `scrape_generic` — Jupyter, HTML, OpenAPI, AsciiDoc, PPTX, RSS, manpage, Confluence, Notion, chat\n\n### Post-processing\n- `enhance_skill` — AI-powered skill enhancement\n- `package_skill` — Package for target platform\n- `upload_skill` — Upload to platform API\n- `install_skill` — End-to-end install workflow\n\n### Advanced\n- `detect_patterns` — Design pattern detection in code\n- `extract_test_examples` — Extract usage examples from tests\n- `build_how_to_guides` — Generate how-to guides from tests\n- `split_config` — Split large configs into focused skills\n- `export_to_weaviate`, `export_to_chroma`, `export_to_faiss`, `export_to_qdrant` — Vector DB export\n\n## CLI Fallback (MCP server not connected)\n\nThe same pipeline is available from the command line (requires `pip install skill-seekers`). Run it with the Bash tool:\n\n```bash\nskill-seekers create <source>                      # auto-detects: URL, owner/repo, ./path, file.pdf, video URL, ...\nskill-seekers package <skill_dir> --target claude  # or gemini/openai/langchain/chroma/...\n```\n\n`create` covers detection, scraping, and building in one step; add `--enhance-level 0` to skip AI enhancement. After it finishes, read the generated `SKILL.md` and summarize what was created.","author":"@yusufkaraaslan","ownerProfile":null,"authorContacts":null,"sourceUrl":"https://github.com/yusufkaraaslan/Skill_Seekers/tree/development/distribution/claude-plugin/skills/skill-builder","license":"MIT","category":"document","lang":"en","tokens":1068,"stars":0,"calls30d":1,"claimed":false,"visibility":"public","origin":"crawler","version":"0.1.0","createdAt":"2026-08-22","updatedAt":"2026-08-22","files":[],"requires":{"mcp":[],"tools":[]},"safety":{"flags":[],"scannedAt":"2026-08-22","hasScripts":false,"networkEndpoints":[]}}