Firecrawl's official routing skill for its CLI: lets an agent search the web, scrape and map sites, crawl sections, monitor pages for changes and interact with pages behind clicks or logins, saving clean markdown to .firecrawl/.
Firecrawl Scrape Skill
Firecrawl's skill for reading known web pages with its CLI: scrape a URL to markdown, links or schema-typed JSON, batch-scrape URL lists, cap PDF pages, and discover and run data-provider tools for structured results.
Install
Other ways to install
About
- Scrape a URL to markdown, full page, links or several formats at once
- Schema-typed JSON extraction with --schema or --schema-file
- batch-scrape jobs for URL lists, with webhooks and per-URL schemas
- Discover, inspect and execute Firecrawl data-provider tools by contract
- PDF page budgets with --max-pages and large-result recovery by job ID
- Saves output to .firecrawl/ files for grep-and-read workflows
Firecrawl Scrape is one of the focused skills shipped in Firecrawl's own CLI repository, next to the umbrella firecrawl skill and siblings for search, interaction and downloads. It covers the moment an agent already knows which page or tool it wants and needs the content: it runs firecrawl scrape and saves the result to a file under .firecrawl/ instead of pasting it into the conversation.
The quick-start covers the common cases: main-content markdown by default or the full page with --full-page, waiting for JavaScript with --wait-for, several URLs at once, markdown and links together, and a targeted --query question about a page. For typed data it passes a JSON schema (inline or from a file) with the json format. For longer lists it moves to batch-scrape, which starts a job you check later, reads URLs from a file and can apply one schema to every page or call a webhook.
Beyond plain pages, the skill explains Firecrawl's catalogue of data-provider tools: discover them with search and list, read the selected tool's contract (required inputs, examples, response key, pagination), then execute it through scrape with exactly the declared fields, checking each result for errors. It also covers recovering large results by job ID rather than repeating a paid call, and limiting PDFs with --max-pages, since PDFs cost one credit per parsed page. Tips favour scraping to a file and grepping it over --query, and escalating to interact when a page needs clicks or forms. It requires the Firecrawl CLI (firecrawl or npx firecrawl-cli) and a Firecrawl account. ISC licensed.
SKILL.md
The instructions your agent loads, as published by Firecrawl.
Related Skills
See all Firecrawl Scrape alternativesSkill: Firecrawl Build: Scrape
by Firecrawl
Integrate Firecrawl's /scrape endpoint into product code — turn a known URL into markdown, HTML, links, metadata or screenshots, with sensible defaults baked in.
Skill: PaddleOCR Document Parsing
by PaddlePaddle
Turn PDFs and scanned images into structured Markdown/JSON: cell-level tables, LaTeX formulas, correct multi-column order.
Skill: PDF
by Anthropic
Read, create, and manipulate PDF files — extract text and tables, merge, split, watermark, fill forms, encrypt, and OCR scanned pages.