firecrawl-reference-architecture（Firecrawl 参考架构）

FireCrawl Reference Architecture

Overview

Production architecture for web scraping and data extraction with FireCrawl. Covers crawl job orchestration, content extraction pipelines, structured data output, and site mapping workflows.

Prerequisites

• FireCrawl API key
• @mendable/firecrawl-js SDK installed
• Queue system for large crawl jobs (optional)
• Storage for extracted content

Architecture Diagram

┌─────────────────────────────────────────────────────┐
│                 Crawl Orchestrator                    │
│  ┌──────────┐  ┌──────────┐  ┌──────────────────┐   │
│  │ Scrape   │  │ Crawl    │  │ Map              │   │
│  │ (single) │  │ (multi)  │  │ (discovery)      │   │
│  └────┬─────┘  └────┬─────┘  └────────┬─────────┘   │
│       │              │                 │             │
│       ▼              ▼                 ▼             │
│  ┌──────────────────────────────────────────────┐    │
│  │         Content Processing Pipeline           │    │
│  │  Markdown │ HTML │ Screenshots │ Structured  │    │
│  └──────────────────────┬───────────────────────┘    │
│                         │                            │
│  ┌──────────────────────┴───────────────────────┐    │
│  │         Output & Storage                      │    │
│  │  JSON Files │ Database │ Vector Store │ S3    │    │
│  └──────────────────────────────────────────────┘    │
└─────────────────────────────────────────────────────┘

Instructions

Step 1: FireCrawl Service Layer

import FirecrawlApp from '@mendable/firecrawl-js';

const firecrawl = new FirecrawlApp({
  apiKey: process.env.FIRECRAWL_API_KEY!,
});

// Single page scrape with markdown output
async function scrapePage(url: string) {
  return firecrawl.scrapeUrl(url, {
    formats: ['markdown', 'html'],
    onlyMainContent: true,
    waitFor: 2000, // Wait for dynamic content  # 2000: 2 seconds in ms
  });
}

// Structured data extraction with schema
async function extractStructured(url: string, schema: any) {
  return firecrawl.scrapeUrl(url, {
    formats: ['extract'],
    extract: {
      schema,
      systemPrompt: 'Extract data precisely according to the schema.',
    },
  });
}

Step 2: Multi-Page Crawl Pipeline

async function crawlSite(baseUrl: string, options?: {
  maxPages?: number;
  includePaths?: string[];
  excludePaths?: string[];
}) {
  const crawlResult = await firecrawl.crawlUrl(baseUrl, {
    limit: options?.maxPages || 50,
    includePaths: options?.includePaths,
    excludePaths: options?.excludePaths || ['/blog/*', '/news/*'],
    scrapeOptions: {
      formats: ['markdown'],
      onlyMainContent: true,
    },
  });

  return crawlResult;
}

// Async crawl for large sites
async function asyncCrawl(baseUrl: string) {
  const job = await firecrawl.asyncCrawlUrl(baseUrl, {
    limit: 500,  # HTTP 500 Internal Server Error
    scrapeOptions: { formats: ['markdown'] },
  });

  // Poll for completion
  let status = await firecrawl.checkCrawlStatus(job.id);
  while (status.status === 'scraping') {
    await new Promise(r => setTimeout(r, 5000));  # 5000: 5 seconds in ms
    status = await firecrawl.checkCrawlStatus(job.id);
  }

  return status;
}

Step 3: Site Map Discovery

async function discoverSiteStructure(url: string) {
  const mapResult = await firecrawl.mapUrl(url);

  // Categorize discovered URLs
  const structure = {
    pages: mapResult.links?.filter(l => !l.includes('/api/')) || [],
    apiDocs: mapResult.links?.filter(l => l.includes('/api/') || l.includes('/docs/')) || [],
    blog: mapResult.links?.filter(l => l.includes('/blog/')) || [],
    total: mapResult.links?.length || 0,
  };

  return structure;
}

Step 4: Structured Extraction Pipeline

import { z } from 'zod';

const ProductSchema = z.object({
  name: z.string(),
  price: z.number(),
  description: z.string(),
  features: z.array(z.string()),
  availability: z.enum(['in_stock', 'out_of_stock', 'preorder']),
});

async function extractProducts(urls: string[]) {
  const results = [];
  for (const url of urls) {
    const data = await extractStructured(url, {
      type: 'object',
      properties: {
        name: { type: 'string' },
        price: { type: 'number' },
        description: { type: 'string' },
        features: { type: 'array', items: { type: 'string' } },
        availability: { type: 'string', enum: ['in_stock', 'out_of_stock', 'preorder'] },
      },
      required: ['name', 'price'],
    });
    results.push(data);
  }
  return results;
}

Error Handling

Examples

Documentation Scraper

async function scrapeDocumentation(docsUrl: string) {
  const sitemap = await discoverSiteStructure(docsUrl);
  const docPages = sitemap.apiDocs.slice(0, 100);

  const crawl = await crawlSite(docsUrl, {
    maxPages: 100,
    includePaths: ['/docs/*', '/api/*', '/guide/*'],
  });

  return crawl;
}

Resources

• FireCrawl Documentation
• FireCrawl Scraping Guide
• FireCrawl Extraction

Output

• Configuration files or code changes applied to the project
• Validation report confirming correct implementation
• Summary of changes made and their rationale