firecrawl-build-crawl
针对文档站点、博客和帮助中心进行大规模爬取，支持自定义路径边界和深度控制的批量数据提取
npx skills add https://github.com/firecrawl/skills --skill firecrawl-build-crawlBefore / After 效果对比
1 组手动逐个访问目标网站页面，复制粘贴内容到本地文件，难以控制爬取深度和路径范围，一个100页的文档站点需要2小时完成收集
配置域名和路径规则后自动批量爬取，智能识别站点结构并按深度边界提取内容，5分钟完成整个站点段的批量下载和结构化存储
firecrawl-build-crawl
Firecrawl Build Crawl
Use this when the product needs many pages from the same site, not just one.
Use This When
-
the feature targets a docs section, blog, or help center
-
you need bulk extraction with path or depth boundaries
-
the product should ingest or refresh a site section at once
Guidance
-
Keep crawl scope narrow with path, depth, or page-count constraints.
-
Prefer
/crawlover many manual/scrapecalls when the site structure is stable. -
Treat this as secondary to
/scrape,/search, and/interactunless the feature is clearly site-wide.
When Not To Use It
-
If you only need one known URL, use firecrawl-build-scrape.
-
If you first need to discover URLs on a single site, start with firecrawl-build-map.
Docs (Source of Truth)
Read the docs for request/response schemas, parameters, and SDK examples before writing integration code:
See Also
Weekly Installs554Repositoryfirecrawl/skillsGitHub Stars2First Seen5 days agoSecurity AuditsGen Agent Trust HubPassSocketPassSnykWarnInstalled onopencode550antigravity550github-copilot550codex550amp550cline550
用户评价 (0)
发表评价
暂无评价
统计数据
用户评分
为此 Skill 评分