Firecrawl

What it does

Firecrawl takes a URL or a search query and returns web content in an LLM-ready form. Scrape and crawl return clean markdown by default, with raw HTML, screenshots, page metadata, and structured JSON available through schemas. Search returns ranked results with full page content, and the Developer Index returns ranked READMEs, issues, merged pull requests, documentation pages, and OpenAPI specs with the matching passages included.

Developers use it to feed retrieval-augmented generation pipelines, deep research agents, lead enrichment, and price monitoring. Firecrawl is open source and is used through its REST API, official SDKs for Python, Node.js, Go, Rust, Java, and Elixir, a CLI, or an MCP server that connects tools such as Claude Code, Cursor, and Codex. Free accounts receive 1,000 credits each month without a card.

Firecrawl suits developers and teams building AI applications that need live web data. Its paid plans are self-serve, with monthly credit allowances and pay-as-you-go top-ups capped by a monthly limit you set. Accounts require signup, and usage consumes credits per page, per search, or per browser minute. Results depend on what the source pages expose, so teams should validate extracted data for their own use case.

What it can do

• Search the web and return full content from results
• Scrape a page into LLM-ready markdown, JSON, HTML, or screenshots
• Crawl and map websites
• Scrape a page and then interact with it using AI prompts or code
• Structured extraction with a JSON schema on scrape
• Developer Index search over READMEs, issues, pull requests, docs and OpenAPI specs

Works with

• OpenAI
• Claude
• Cursor
• Claude Code
• Codex
• Windsurf
• Model Context Protocol (MCP) server

Example workflows

Scrape a page into LLM-ready markdown

Illustrative example. Input: a URL such as https://example.com and an API key. Steps: call the scrape method with the URL; receive the page as markdown, optionally with a JSON schema for structured fields. Output: markdown or structured JSON for use in an LLM prompt or RAG index. Each extracted page costs 1 credit.

Find the fix for a coding error

Illustrative example. Input: a natural-language error description such as 'why is my retry backoff not firing on 429'. Steps: send the query to the Developer Index search endpoint with types set to issue and pull_request; read the ranked results and matched passages. Output: a ranked list of related issues, pull requests and documentation with the matching text. Search costs 2 credits per 10 results.

Before you choose

• Credits are consumed per request: scrape, crawl and map cost 1 credit per page; search costs 2 credits per 10 results; interact costs 2 credits per browser minute; advanced formats add credits.
• Credits roll over only on Scale and Enterprise; Hobby, Standard and Growth credits do not roll over.
• Pay-as-you-go needs a paid plan and charges in 5 USD increments up to a monthly limit you set; the Agent feature is in preview with dynamic pricing.

Listed under

• Coding
• AI agents
• Crawler
• Developer tools
• LLM data
• Open source
• Search API

Get started

1,000 free credits every month, no card required.

1. Sign up at https://www.firecrawl.dev/signin?view=signup and copy your API key from the dashboard.
2. Install an SDK (for example pip install firecrawl-py) or the CLI with npx -y firecrawl-cli@latest, then authenticate with your API key.
3. Make a first call such as scrape or search, or for coding agents run npx -y firecrawl-cli@latest setup developer-index to connect the Developer Index.

Paid plans are billed monthly or annually; the annual figures shown are per month billed annually. Usage beyond plan credits is charged via pay-as-you-go in 5 USD increments up to a monthly limit you set. Marketplace purchases (AWS, Google Cloud) are billed through those marketplaces. Annual figures are displayed monthly equivalents; confirm the full annual checkout total. Scrape, Crawl and Map use one credit/page; Search uses two credits/ten results; Interact uses two credits/browser minute. Rollover is limited by plan, including one month on Scale; lower plans do not roll over.

Source: firecrawl.dev/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Firecrawl | Web Data API for AI Agents ↗www.firecrawl.dev
• Pricing | Firecrawl ↗www.firecrawl.dev
• Developer Index: Code & Docs Search API | Firecrawl ↗www.firecrawl.dev

Company website

firecrawl.dev: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.