firecrawl-scrape — Claude Code skill: read pages, pull data
The skill reads a known page into markdown or JSON and runs discovered data-provider tools. Useful once the URL is already picked.
npx skills add firecrawl/cli --skill firecrawl-scrape -a claude-codeWhat firecrawl-scrape does
The skill exposes firecrawl scrape and npx firecrawl-cli to the agent. A basic call pulls a page into markdown and writes it to .firecrawl/page.md. The --only-main-content flag drops nav and footer, --wait-for 3000 waits for JavaScript to render. Several URLs can go in one call, but then each file lands in .firecrawl/ and -o is ignored.
For typed fields instead of markdown, pass a JSON schema: --format json --schema '{...}'. A large schema is easier to keep in a file and load with --schema-file. The result sits under the .json key. The schema applies only when json is in --format; add other formats with commas.
A list of known URLs goes through firecrawl batch-scrape. Reach for it when you need formats beyond markdown, JSON extraction, a webhook, or a job to check later. The run prints a job ID, and you check the job by that ID afterwards. PDFs are billed at 1 credit per parsed page, so --max-pages takes an integer from 1 to 10000. Save large output with --json -o and inspect it in sections; keep stderr away from JSON stdout rather than merging with 2>&1.
What it helps with
- read a page at a known URL as markdown
- pull a plan price off a page with
--query - get typed fields from a JSON schema
- process a list of known URLs in a batch
- cap PDF parsing by page count
Example prompts for the agent
- Read
https://example.com/pricingand save the markdown to.firecrawl/page.md - Scrape only the main content, no nav or footer
- Extract the
planandpricefields using the schema inschema.json
How to install firecrawl-scrape
1Skills CLI (Vercel)
npx skills add firecrawl/cli --skill firecrawl-scrape -a claude-code2Manually: put the files in place
~/.claude/skills/firecrawl-scrape/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/firecrawl-scrape/
1Skills CLI (Vercel)
npx skills add firecrawl/cli --skill firecrawl-scrape -a codex2Manually: put the files in place
.agents/skills/firecrawl-scrape/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add firecrawl/cli --skill firecrawl-scrape -a cursor2Manually: put the files in place
.cursor/skills/firecrawl-scrape/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install firecrawl/cli firecrawl-scrape2Manually: put the files in place
.github/skills/firecrawl-scrape/SKILL.md1Skills CLI (Vercel)
npx skills add firecrawl/cli --skill firecrawl-scrape -a gemini-cli2Platform CLI
gemini skills install https://github.com/firecrawl/cli --path skills/firecrawl-scrapeManually: put the files in place
.devin/skills/firecrawl-scrape/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add firecrawl/cli --skill firecrawl-scrape -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about firecrawl-scrape
How do I install firecrawl-scrape in Claude Code?
Run npx skills add firecrawl/cli --skill firecrawl-scrape -a claude-code. The skill lands in ~/.claude/skills/firecrawl-scrape/SKILL.md.
What does firecrawl-scrape do?
It reads a known page into markdown or JSON and executes a selected data-provider tool for structured results. Discovery and contract inspection happen through search and list before execution.
How is scrape different from batch-scrape?
scrape handles one page or a few in markdown. batch-scrape is for a list of URLs when you need other formats, JSON extraction, a webhook, or a job you check later.
