菜单
Use when scraping docs websites into local markdown files, crawling docs, or building AI-readable docs context.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when setting up or reviewing a development environment, especially Varlock env schemas, gitignored env files, macOS Keychain/local secret storage, SOPS/age optional GitOps secrets, CI secret access, dotenv bootstrapping, or secure local dev onboarding.
Use when the repo has .beans or the user mentions beans, beans-prime, flat-file issues, task tracking, or issue status changes.
Opinionated JB GitHub Actions release-attempt workflow. Use only when the repo already has .github/workflows/jb-release-v1.yaml (or equivalent JB release-attempt workflow) or the user explicitly asks to set it up. Do not use for ordinary local releases; use jb-local-release instead.
Use when the user asks to run a local release flow: prepare version/changelog changes, run local checks, commit, tag, and optionally publish only when explicitly requested.
Query MDN Web Docs through the official MDN MCP server using a persistent mcporter config, avoiding a direct MCP connection in the agent harness. Use when checking current web platform docs, CSS/HTML/JavaScript/Web API behavior, MDN browser compatibility data, Baseline support, or when the user asks whether a browser feature is supported.
Use when the user asks for Chrome MCP/DevTools MCP via mcporter, existing Chrome tabs, console/network inspection, screenshots, or performance traces.
| name | jb-docs-scraper |
| description | Use when scraping docs websites into local markdown files, crawling docs, or building AI-readable docs context. |
Scrape any documentation website into local markdown files. Uses crawl4ai for async web crawling.
# Scrape any documentation URL
uv run --with crawl4ai python ./references/scrape_docs.py <URL>
# Examples
uv run --with crawl4ai python ./references/scrape_docs.py https://mediasoup.org/documentation/v3/
uv run --with crawl4ai python ./references/scrape_docs.py https://docs.rombo.co/tailwind
Output goes to ./docs/<auto-detected-name>/ by default.
uv run --with crawl4ai playwright install
uv run --with crawl4ai python ./references/scrape_docs.py <URL> [OPTIONS]
| Option | Description | Default |
|---|---|---|
-o, --output PATH | Output directory | ./docs/<auto-detected-name> |
--max-depth N | Maximum link depth | 6 |
--max-pages N | Maximum pages to scrape | 500 |
--url-pattern PATTERN | URL filter (glob) | Auto-detected |
-q, --quiet | Suppress verbose output | False |
# Basic - scrape to ./docs/documentation_v3/
uv run --with crawl4ai python ./references/scrape_docs.py \
https://mediasoup.org/documentation/v3/
# Custom output directory
uv run --with crawl4ai python ./references/scrape_docs.py \
https://docs.rombo.co/tailwind \
--output ./my-tailwind-docs
# Limit crawl scope
uv run --with crawl4ai python ./references/scrape_docs.py \
https://tanstack.com/start/latest/docs/framework/react/overview \
--max-pages 50 \
--max-depth 3
# Custom URL pattern filter
uv run --with crawl4ai python ./references/scrape_docs.py \
https://example.com/docs/api/v2/ \
--url-pattern "*api/v2/*"
docs/<name>/
index.md # Root page
getting-started.md
api/
overview.md
client.md
guides/
installation.md
| Issue | Solution |
|---|---|
Playwright browser binaries are missing | Run uv run --with crawl4ai playwright install |
| Empty output | Check if URL pattern matches actual doc URLs. Try --url-pattern |
| Missing pages | Increase --max-depth or --max-pages |
| Wrong pages scraped | Use stricter --url-pattern |
--max-pages 10 to verify config before full crawl