firecrawl

Automate web scraping, embedding, and retrieval with the Firecrawl CLI.

69|7|Updated Feb 8, 2026
One-click install
npx skills add https://github.com/jmagar/claude-homelab --skill firecrawl-jmagar
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: firecrawl
Source: https://github.com/jmagar/claude-homelab/tree/main/skills/firecrawl
Command: npx skills add https://github.com/jmagar/claude-homelab --skill firecrawl-jmagar

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq, md5sum, sed, grep, awk, cut, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Firecrawl replaces scattered web data gathering tools by providing an end-to-end workflow for extracting, embedding, and retrieving web content via a CLI, delivering clean Markdown optimized for LLM context windows.

Core Features & Use Cases

  • Scrape single pages to LLMed Markdown, HTML, or links for context windows
  • Crawl entire sites to index content and enable semantic search
  • Map website structures to build sitemaps and discovery datasets
  • Use cases include building a knowledge base, monitoring docs, and performing research with retrieval

Quick Start

Install the Firecrawl CLI, configure credentials, and run a test scrape to index content.

Frequently Asked Questions about firecrawl

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract web content for LLM context windows and semantic search?

You can extract web content for LLM context windows by scraping single pages into clean Markdown or HTML. This process automates embedding and retrieval operations to build structured knowledge bases suitable for semantic search.

What is the best way to crawl an entire website to build a vector database?

The best way to crawl an entire website to build a vector database is by indexing content across large web sources. This workflow maps site structures, extracts documentation, and generates embeddings to enable semantic retrieval.

Do I need the Firecrawl CLI to automate documentation harvesting and embedding?

Yes, automating documentation harvesting and embedding requires the Firecrawl CLI. It enforces secure configuration and parameter validation to deliver structured outputs suitable for LLM processing within research pipelines.

Can I map website structures to build sitemaps and discovery datasets?

Yes, you can map website structures to build sitemaps and discovery datasets. This functionality extracts links and indexes content to support research pipelines and knowledge-base building operations.

Does web scraping for RAG work with standard shell utilities like jq and grep?

Yes, web scraping for RAG works with standard shell utilities like jq, sed, and grep. These dependencies process structured outputs and validate parameters during data extraction and embedding workflows.

Why use Markdown extraction instead of raw HTML for retrieval pipelines?

Markdown extraction is used instead of raw HTML for retrieval pipelines because it delivers clean text optimized for LLM context windows. This reduces noise and improves embedding accuracy for semantic search queries.