bextuychiev/firecrawl-claude-code-skill

Overview

This skill provides programmatic web access via the Firecrawl CLI wrapper to fetch page content, capture screenshots, extract structured data, run web searches, and crawl documentation sites. It converts pages to clean Markdown, supports targeted JSON-schema extraction, and can save full-page screenshots. Use it when you need up-to-date web information, scraped data, or an offline snapshot of docs or pages.

How this skill works

The skill runs a bundled Python script that calls the Firecrawl API to load pages, render content, and return results in the requested format. Commands include markdown conversion (optionally main-content only), full-page screenshots, schema-driven JSON extraction with optional prompts for accuracy, web search with result limits, and site crawling with output directory and page limits. Each crawled page consumes a credit, so the tool exposes limits and output options to control cost.

When to use it

• You need current information from the live web or recent documentation.
• You want a clean Markdown version of a webpage for notes or content ingestion.
• You need structured data extracted from product pages, posts, or listings.
• You want a full-page screenshot for visual review or reporting.
• You need to crawl and archive documentation or a section of a website.

Best practices

• Use --main-only when you only need primary article content to reduce noise.
• Define a JSON schema that matches the fields you need before running extract.
• Include a short extraction prompt to improve field accuracy and context.
• Set reasonable crawl limits to control credits and avoid excessive cost.
• Save crawl output to a directory for offline review and downstream processing.

Example use cases

• Fetch a blog post as Markdown for summarization or citation in a report.
• Extract product title, price, and features using a JSON schema for a price tracker.
• Take a full-page PNG of a landing page for QA or visual change detection.
• Search the web for the latest framework changes and limit results to top sources.
• Crawl a docs site to build a local knowledge base or feed a search index.