Firecrawl Skill Pack

30 Claude Code skills for Firecrawl — turn any website into LLM-ready markdown with scrape, crawl, map, and extract APIs

What It Does

Firecrawl converts websites into clean markdown or structured JSON for LLM consumption. It handles JavaScript rendering, anti-bot bypassing, and pagination automatically. This skill pack teaches Claude Code every Firecrawl pattern from hello-world to enterprise content pipelines.

SDK: @mendable/firecrawl-js (npm) / firecrawl-py (PyPI) API: api.firecrawl.dev/v1 — scrape, crawl, map, extract, batch scrape Pricing: Credit-based (1 credit = 1 page scraped)

Installation

/plugin install firecrawl-pack@claude-code-plugins-plus

Skills (30)

Getting Started (S01-S04)

| Skill | What It Covers | |-------|---------------| | firecrawl-install-auth | Install @mendable/firecrawl-js, configure FIRECRAWL_API_KEY, verify connection | | firecrawl-hello-world | Four minimal examples: scrape, crawl, map, extract | | firecrawl-local-dev-loop | Self-hosted Docker for zero-credit dev, vitest mocks, integration tests | | firecrawl-sdk-patterns | Singleton client, typed wrappers, retry with backoff, Zod validation |

Core Workflows (S05-S06)

| Skill | What It Covers | |-------|---------------| | firecrawl-core-workflow-a | scrapeUrl + crawlUrl + asyncCrawlUrl with polling and content processing | | firecrawl-core-workflow-b | extract with JSON schemas, batchScrapeUrls, mapUrl + selective scrape |

Operations (S07-S12)

| Skill | What It Covers | |-------|---------------| | firecrawl-common-errors | Fix 401/402/429/500 errors, empty markdown, stuck crawls, wrong package name | | firecrawl-debug-bundle | Collect SDK version, API health, credit balance, diagnostic scrape for support | | firecrawl-rate-limits | Exponential backoff, p-queue concurrency, proactive throttling, batch endpoints | | firecrawl-security-basics | API key storage, webhook HMAC-SHA256 verification, key rotation, content sanitization | | firecrawl-prod-checklist | Pre-deploy validation: credentials, crawl limits, error handling, monitoring, rollback | | firecrawl-upgrade-migration | v0 to v1/v2 migration: crawlerOptions to flat options, new methods, breaking changes |

Pro (P13-P18)

| Skill | What It Covers | |-------|---------------| | firecrawl-ci-integration | GitHub Actions workflow, integration tests, mock-based unit tests, credit-aware CI | | firecrawl-deploy-integration | Vercel serverless, Cloud Run, self-hosted Docker Compose, webhook endpoints | | firecrawl-webhooks-events | crawl.started/page/completed events, signature verification, polling fallback | | firecrawl-performance-tuning | Format selection, waitFor tuning, LRU caching, batch vs individual, map-first pattern | | firecrawl-cost-tuning | Crawl limits, map+selective scrape, caching, credit monitoring, budget enforcement | | firecrawl-reference-architecture | Scrape/crawl/map/extract pipeline, content processing, chunking, storage patterns |

Flagship (F19-F24)

| Skill | What It Covers | |-------|---------------| | firecrawl-multi-env-setup | Dev/staging/prod config, self-hosted Docker for dev, credit-safe wrappers | | firecrawl-observability | Scrape metrics, credit burn tracking, content quality scoring, Prometheus alerts | | firecrawl-incident-runbook | Quick triage, decision tree by error code, communication templates, postmortem | | firecrawl-data-handling | Markdown cleaning, Zod-validated extraction, deduplication, RAG chunking, storage | | firecrawl-enterprise-rbac | Per-team API keys, domain allowlists, credit budgets, gateway proxy pattern | | firecrawl-migration-deep-dive | Migrate from Puppeteer/Playwright/Cheerio to Firecrawl with adapter pattern |

Flagship+ (X25-X30)

| Skill | What It Covers | |-------|---------------| | firecrawl-advanced-troubleshooting | Layer-by-layer isolation, empty scrape debugging, screenshot diagnosis, timing analysis | | firecrawl-load-scale | Throughput measurement, batch scaling, p-queue workers, parallel async crawls | | firecrawl-reliability-patterns | Circuit breaker, crawl-to-scrape fallback, content validation, credit guard | | firecrawl-policy-guardrails | Domain blocklists, credit budgets, content quality gates, per-domain rate limits | | firecrawl-architecture-variants | On-demand vs scheduled pipeline vs real-time RAG ingestion decision guide | | firecrawl-known-pitfalls | 10 anti-patterns with BAD/GOOD code: unbounded crawl, wrong import, no validation |

Quick Start

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

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

// Scrape a page to markdown
const page = await firecrawl.scrapeUrl("https://docs.example.com", {
  formats: ["markdown"],
  onlyMainContent: true,
});

// Crawl an entire site
const site = await firecrawl.crawlUrl("https://docs.example.com", {
  limit: 50,
  scrapeOptions: { formats: ["markdown"] },
});

// Discover all URLs instantly
const map = await firecrawl.mapUrl("https://docs.example.com");

// Extract structured data with LLM
const data = await firecrawl.scrapeUrl("https://example.com/pricing", {
  formats: ["extract"],
  extract: { schema: { type: "object", properties: { plans: { type: "array" } } } },
});

Key Links

• Firecrawl Docs
• Firecrawl Dashboard
• Node SDK on npm
• GitHub

License

MIT