Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
benedictking avatar

Firecrawl Scraper

  • 523 installs
  • 14 repo stars
  • Updated August 1, 2026
  • benedictking/benedictking-skills

Firecrawl Scraper is a Claude Code skill that routes coding agents to Firecrawl API endpoints for scraping, crawling, screenshots, PDF parsing, and structured extraction when developers need reliable web data inside agen

About

Firecrawl Scraper is a MIT-licensed agent skill (version 1.0.2) that teaches Claude Code when and how to call Firecrawl API endpoints for single-page scrape, full-site crawl, batch URL processing, screenshots, and PDF parsing. The skill selects the correct endpoint from user intent—markdown, HTML, JSON, screenshot, or PDF output—and runs via Bash with network access to Firecrawl. Developers reach for Firecrawl Scraper when agents must fetch real page content, crawl documentation sites, or extract structured data without hand-writing scraper scripts or hallucinating page content. It is user-invocable and designed for Claude Code with Node.js as a runtime dependency.

  • Supports scrape, crawl, map, batch-scrape, crawl-status, batch-status and batch-errors endpoints
  • Two-phase architecture that isolates HTTP execution in a sub-skill to reduce token usage
  • Automatically selects the correct Firecrawl endpoint based on user intent
  • Returns clean markdown, HTML, JSON, screenshots or structured data from any web target
  • Designed for Claude Code with Node.js and network access

Firecrawl Scraper by the numbers

  • 523 all-time installs (skills.sh)
  • Ranked #1,717 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
  • Data as of Aug 2, 2026 (Skillselion catalog sync)
npx skills add https://github.com/benedictking/benedictking-skills --skill firecrawl-scraper

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs523
repo stars14
Last updatedAugust 1, 2026
Repositorybenedictking/benedictking-skills

How do agents scrape pages with Firecrawl?

Let their coding agent scrape websites, crawl pages, extract structured data, capture screenshots, or parse PDFs using the Firecrawl API.

Who is it for?

Developers who want Claude Code agents to fetch, crawl, or structure live web content through Firecrawl without writing custom scraper code.

Skip if: Developers who need offline scraping, browser automation without Firecrawl, or projects blocked from external API network calls.

When should I use this skill?

User asks to scrape a page, crawl a website, batch-scrape URLs, capture a screenshot, parse a PDF, or extract structured data from the web.

What you get

Markdown, HTML, JSON extracts, screenshots, PDF parses, and crawl result sets from target URLs.

  • Scraped page content
  • Crawl datasets
  • Screenshots and PDF extracts

By the numbers

  • Skill version 1.0.2
  • Supports scrape, crawl, and batch-scrape Firecrawl endpoints

Files

SKILL.mdMarkdownGitHub ↗

Firecrawl Scraper Skill

Trigger Conditions & Endpoint Selection

Choose Firecrawl endpoint based on user intent:

  • scrape: Need to extract content from a single web page (markdown, html, json, screenshot, pdf)
  • crawl: Need to crawl entire website with depth control and path filtering
  • map: Need to quickly get a list of all URLs on a website
  • batch-scrape: Need to scrape multiple URLs in parallel
  • crawl-status: Given crawl job ID, check crawl progress/results (optional --wait)
  • batch-status: Given batch job ID, check batch scrape progress/results (optional --wait)
  • batch-errors: Given batch job ID, retrieve batch scrape errors

Recommended Architecture (Main Skill + Sub-skill)

This skill uses a two-phase architecture:

1. Main skill (current context): Understand user question → Choose endpoint → Assemble JSON payload 2. Sub-skill (fork context): Only responsible for HTTP call execution, avoiding conversation history token waste

Execution Method

Use Task tool to invoke firecrawl-fetcher sub-skill, passing command and JSON (stdin):

Task parameters:
- subagent_type: Bash
- description: "Call Firecrawl API"
- prompt: cat <<'JSON' | node scripts/firecrawl-api.cjs <scrape|crawl|map|batch-scrape|crawl-status|batch-status|batch-errors> [--wait]
  { ...payload... }
  JSON

Payload Examples

1) Scrape Single Page

cat <<'JSON' | node scripts/firecrawl-api.cjs scrape
{
  "url": "https://example.com",
  "formats": ["markdown", "links"],
  "onlyMainContent": true,
  "includeTags": [],
  "excludeTags": ["nav", "footer"],
  "waitFor": 0,
  "timeout": 30000
}
JSON

Available formats:

  • "markdown", "html", "rawHtml", "links", "images", "summary"
  • {"type": "json", "prompt": "Extract product info", "schema": {...}}
  • {"type": "screenshot", "fullPage": true, "quality": 85}

2) Scrape with Actions (Page Interaction)

cat <<'JSON' | node scripts/firecrawl-api.cjs scrape
{
  "url": "https://example.com",
  "formats": ["markdown"],
  "actions": [
    {"type": "wait", "milliseconds": 2000},
    {"type": "click", "selector": "#load-more"},
    {"type": "wait", "milliseconds": 1000},
    {"type": "scroll", "direction": "down", "amount": 500}
  ]
}
JSON

Available actions:

  • wait, click, write, press, scroll, screenshot, scrape, executeJavascript

3) Parse PDF

cat <<'JSON' | node scripts/firecrawl-api.cjs scrape
{
  "url": "https://example.com/document.pdf",
  "formats": ["markdown"],
  "parsers": ["pdf"]
}
JSON

4) Extract Structured JSON

cat <<'JSON' | node scripts/firecrawl-api.cjs scrape
{
  "url": "https://example.com/product",
  "formats": [
    {
      "type": "json",
      "prompt": "Extract product information",
      "schema": {
        "type": "object",
        "properties": {
          "name": {"type": "string"},
          "price": {"type": "number"},
          "description": {"type": "string"}
        },
        "required": ["name", "price"]
      }
    }
  ]
}
JSON

5) Crawl Entire Website

cat <<'JSON' | node scripts/firecrawl-api.cjs crawl
{
  "url": "https://docs.example.com",
  "formats": ["markdown"],
  "includePaths": ["^/docs/.*"],
  "excludePaths": ["^/blog/.*"],
  "maxDiscoveryDepth": 3,
  "limit": 100,
  "allowExternalLinks": false,
  "allowSubdomains": false
}
JSON

5.1) Crawl + Wait for Completion

cat <<'JSON' | node scripts/firecrawl-api.cjs crawl --wait
{
  "url": "https://docs.example.com",
  "formats": ["markdown"],
  "limit": 100
}
JSON

6) Map Website URLs

cat <<'JSON' | node scripts/firecrawl-api.cjs map
{
  "url": "https://example.com",
  "search": "documentation",
  "limit": 5000
}
JSON

7) Batch Scrape Multiple URLs

cat <<'JSON' | node scripts/firecrawl-api.cjs batch-scrape
{
  "urls": [
    "https://example.com/page1",
    "https://example.com/page2",
    "https://example.com/page3"
  ],
  "formats": ["markdown"]
}
JSON

Returns async job response: { "success": true, "id": "<batch-id>", "url": "..." }

7.1) Batch Scrape + Wait for Completion

cat <<'JSON' | node scripts/firecrawl-api.cjs batch-scrape --wait
{
  "urls": [
    "https://example.com/page1",
    "https://example.com/page2"
  ],
  "formats": ["markdown"]
}
JSON

7.2) Check Batch Scrape Status

node scripts/firecrawl-api.cjs batch-status <batch-id>

Wait for completion:

node scripts/firecrawl-api.cjs batch-status <batch-id> --wait

7.3) Get Batch Scrape Errors

node scripts/firecrawl-api.cjs batch-errors <batch-id>

8) Check Crawl Status

node scripts/firecrawl-api.cjs crawl-status <crawl-id>

Wait for completion:

node scripts/firecrawl-api.cjs crawl-status <crawl-id> --wait

Key Features

Formats

  • markdown: Clean markdown content
  • html: Parsed HTML
  • rawHtml: Original HTML
  • links: All links on page
  • images: All images on page
  • summary: AI-generated summary
  • json: Structured data extraction with schema
  • screenshot: Page screenshot (PNG)

Content Control

  • onlyMainContent: Extract only main content (default: true)
  • includeTags: CSS selectors to include
  • excludeTags: CSS selectors to exclude
  • waitFor: Wait time before scraping (ms)
  • maxAge: Cache duration (default: 48 hours)

Actions (Browser Automation)

  • wait: Wait for specified time
  • click: Click element by selector
  • write: Input text into field
  • press: Press keyboard key
  • scroll: Scroll page
  • executeJavascript: Run custom JS

Crawl Options

  • includePaths: Regex patterns to include
  • excludePaths: Regex patterns to exclude
  • maxDiscoveryDepth: Maximum crawl depth
  • limit: Maximum pages to crawl
  • allowExternalLinks: Follow external links
  • allowSubdomains: Follow subdomains

Environment Variables & API Key

Two ways to configure API Key (priority: environment variable > .env):

1. Environment variable: FIRECRAWL_API_KEY 2. .env file: Place in .env, can copy from .env.example

Response Format

All endpoints return JSON with:

  • success: Boolean indicating success
  • data: Extracted content (format depends on endpoint)
  • For crawl: Returns job ID, use crawl-status (or GET /v2/crawl/{id}) to check status
  • For batch-scrape: Returns async job response ({ success, id, url }), use batch-status (or GET /v2/batch/scrape/{id}) to poll status
  • Batch status response includes status, total, completed, creditsUsed, expiresAt, next, data[]
  • next: pagination URL for large/incomplete results (script returns raw response; follow manually if needed)

Related skills

How it compares

Choose Firecrawl Scraper when you already use Firecrawl and want endpoint selection baked into agent workflows instead of ad-hoc curl or custom SDK scripts.

FAQ

What Firecrawl endpoints does firecrawl-scraper use?

Firecrawl Scraper maps intent to Firecrawl endpoints: scrape for a single page, crawl for an entire site, and batch operations for multiple URLs. Output formats include markdown, HTML, JSON, screenshots, and PDF parsing depending on the request.

What does firecrawl-scraper require to run?

Firecrawl Scraper requires Claude Code, Node.js, network access to the Firecrawl API, and Bash/Read tool permissions. The skill is MIT-licensed at version 1.0.2 and is marked user-invocable in its manifest.

AI & Agent Buildingagentsautomation

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.