Firecrawl core workflow a
'Execute Firecrawl primary workflow: scrape and crawl websites into LLM-ready markdown.From its SKILL.md
npx -y skills add jeremylongshore/claude-code-plugins-plus-skills --skill firecrawl-core-workflow-aAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
5.8 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it
Firecrawl Core Workflow A — Scrape & Crawl
Overview
Primary workflow for Firecrawl: convert websites into clean LLM-ready markdown. Covers single-page scraping with scrapeUrl, multi-page crawling with crawlUrl, async crawl jobs with polling, and content processing pipelines.
Prerequisites
@mendable/firecrawl-jsinstalledFIRECRAWL_API_KEYenvironment variable set- Target URL(s) identified
Instructions
Step 1: Single-Page Scrape
import FirecrawlApp from "@mendable/firecrawl-js";
const firecrawl = new FirecrawlApp({
apiKey: process.env.FIRECRAWL_API_KEY!,
});
// Scrape a single page to clean markdown
const result = await firecrawl.scrapeUrl("https://docs.example.com/api", {
formats: ["markdown"],
onlyMainContent: true, // strips nav, footer, sidebars
waitFor: 2000, // wait 2s for JS to render
});
if (result.success) {
console.log("Title:", result.metadata?.title);
console.log("Source:", result.metadata?.sourceURL);
console.log("Markdown:", result.markdown?.substring(0, 200));
}
Step 2: Multi-Page Synchronous Crawl
// Crawl a site — Firecrawl follows links, renders JS, returns all pages
const crawlResult = await firecrawl.crawlUrl("https://docs.example.com", {
limit: 50, // max pages to crawl
maxDepth: 3, // link depth from start URL
includePaths: ["/docs/*", "/api/*"], // only these paths
excludePaths: ["/blog/*", "/changelog/*"],
allowBackwardLinks: false, // only crawl child paths
scrapeOptions: {
formats: ["markdown"],
onlyMainContent: true,
},
});
console.log(`Crawled ${crawlResult.data?.length} pages`);
for (const page of crawlResult.data || []) {
console.log(` ${page.metadata?.sourceURL}: ${page.markdown?.length} chars`);
}
Step 3: Async Crawl for Large Sites
// Start an async crawl job — returns immediately with job ID
const job = await firecrawl.asyncCrawlUrl("https://docs.example.com", {
limit: 500,
scrapeOptions: { formats: ["markdown"] },
});
console.log(`Crawl started: ${job.id}`);
// Poll for completion with backoff
let pollInterval = 2000;
let status = await firecrawl.checkCrawlStatus(job.id);
while (status.status === "scraping") {
console.log(`Progress: ${status.completed}/${status.total} pages`);
await new Promise(r => setTimeout(r, pollInterval));
pollInterval = Math.min(pollInterval * 1.5, 30000);
status = await firecrawl.checkCrawlStatus(job.id);
}
if (status.status === "completed") {
console.log(`Done: ${status.data?.length} pages scraped`);
} else {
console.error("Crawl failed:", status.error);
}
Step 4: Process and Store Results
import { writeFileSync, mkdirSync } from "fs";
function processResults(pages: any[], outputDir: string) {
mkdirSync(outputDir, { recursive: true });
const manifest = pages.map((page, i) => {
const url = page.metadata?.sourceURL || `page-${i}`;
const slug = new URL(url).pathname
.replace(/\//g, "_")
.replace(/^_|_$/g, "") || "index";
const filename = `${slug}.md`;
// Clean markdown: collapse whitespace, remove JS links
const content = (page.markdown || "")
.replace(/\n{3,}/g, "\n\n")
.replace(/\[.*?\]\(javascript:.*?\)/g, "")
.trim();
writeFileSync(`${outputDir}/${filename}`, content);
return { url, filename, chars: content.length };
});
writeFileSync(`${outputDir}/manifest.json`, JSON.stringify(manifest, null, 2));
return manifest;
}
Output
- Clean markdown files per crawled page
manifest.jsonwith URL-to-file mapping- Crawl summary with page count and failures
Error Handling
| Error | Cause | Solution |
|---|---|---|
Empty markdown | JS content not rendered | Increase waitFor to 5000ms |
429 Too Many Requests | Rate limit hit | Back off, reduce concurrency |
| Crawl returns few pages | URL filters too strict | Widen includePaths patterns |
402 Payment Required | Credits exhausted | Check balance, reduce limit |
| Partial crawl results | Site blocks bot on some pages | Use scrapeUrl for failed URLs individually |
Examples
Scrape with Multiple Formats
const result = await firecrawl.scrapeUrl("https://example.com", {
formats: ["markdown", "html", "links"],
onlyMainContent: true,
});
console.log("Markdown:", result.markdown?.length);
console.log("HTML:", result.html?.length);
console.log("Links:", result.links?.length);
Crawl with Webhook (No Polling)
const job = await firecrawl.asyncCrawlUrl("https://docs.example.com", {
limit: 100,
scrapeOptions: { formats: ["markdown"] },
webhook: {
url: "https://api.yourapp.com/webhooks/firecrawl",
events: ["completed", "page"],
},
});
console.log(`Crawl ${job.id} started — webhook will fire on completion`);
Resources
Next Steps
For structured data extraction, see firecrawl-core-workflow-b.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.