Available in Playground
Firecrawl Search API
I don’t know the target page
Find relevant pages from a query.
RESULT
Relevant web sources for research
Firecrawl API on SandBaseFirecrawl API for Search, Scraping, and Crawling
Access Firecrawl through a SandBase API key. Search for sources, scrape pages, discover site URLs or crawl collections, then use Markdown and structured data in your application or Agent workflow.
FirecrawlWEB DATA FOR AI One SandBase API key Five Firecrawl capabilities Markdown · Structured data
Firecrawl is a web data API for AI agents and LLMs. It turns live web pages into clean Markdown and structured data.
- Founded 2022
- YC S22
- San Francisco
- Open Source
FIRECRAWL CAPABILITIES
Firecrawl API Capabilities Available on SandBase
Use Search to find pages, Scrape to read a URL, Map to discover site URLs, Batch Scrape to process a URL list, and Crawl to collect content across a site.
WEB → CLEAN CONTENT → AGENTILLUSTRATIVE WORKFLOW
01 / INPUT
A page you want to read.
A page you want to read.
Sample data · no web request02 / CLEAN OUTPUT
# A practical guide to Agent sandboxes ## Keep the work isolated Run each task in a scoped environment. ## Keep the useful output Pass clean content to your Agent.
Ready for Your AgentResearch · summarize · reason
Illustrative Workflow · Agent Integrations Not Officially Verified
Discover the Source Keep the Structure Let the Agent Work
FIRECRAWL ENDPOINTS
Firecrawl API Endpoints
Available in Playground
Firecrawl Scrape API
I have one known URL
Extract clean content from one page.
RESULT
Clean Markdown or structured page data
Available in Playground
Firecrawl Map API
I want to inspect site structure
Discover the URLs that make up a site.
RESULT
A list of discovered site URLs
Available in Playground
Firecrawl Batch Scrape API
I already have a list of URLs
Extract content from a known URL list.
RESULT
Clean content from the supplied URL list
Available in Playground
Firecrawl Crawl API
I need coverage of a site
Follow a site and collect page content.
RESULT
Collected page content across a site
FIRECRAWL ON SANDBASE
Why Use the Firecrawl API Through SandBase?
One SandBase API Key
Access Firecrawl API capabilities with your SandBase API key instead of managing a separate key for each tool.
Shared Balance Across APIs and Models
Pay for Firecrawl API work from the same SandBase balance used across APIs and models, with request and cost visibility.
Test Firecrawl in Playground
Explore Firecrawl Search, Scrape, Map, Batch Scrape, and Crawl in Playground before connecting them to your application.
Connect Firecrawl to Agents Through API or MCP
Use Firecrawl API content with Agent workflow logic, Skills, models, and other data tools through the unified API or SandBase MCP.
FIRECRAWL IN YOUR WORKFLOW
Use the Firecrawl API in an Agent Workflow
Start with Firecrawl API content, then connect it to the workflow logic and model your Agent needs.
01
Firecrawl API
Start with web data
02
Search / Scrape / Crawl
Choose the capability
03
Skill or Workflow Logic
Handle the returned content
04
Agent
Use content for the task
05
Swappable Model
Choose the reasoning
EXAMPLE SKILLS · NOT VERIFIED DEPENDENCIES
Web ScraperURL to MarkdownWebsite Monitor
These labels illustrate possible workflows, not a verified catalog or automatic monitoring service.
Access Firecrawl Through MCP
SandBase MCP is another way to access Firecrawl from your AI tools. Confirm the available tool names and supported capabilities in the docs.
RELATED APIS
APIs Related to Firecrawl
Explore complementary web search, research, extraction, and web-agent APIs. Choose the capability that fits your workflow; these services are not all direct replacements for Firecrawl.
Tavily
AI-oriented web search and research.
Exa
Neural and semantic web search and research.
Jina
A reader, search, and embeddings ecosystem that turns web information into model-ready inputs.
TinyFish
Web agents and structured web data extraction.
Before production, confirm Firecrawl API pricing, request and concurrency limits, data retention, and website access permissions in Playground or the docs. Respect website permissions; do not assume unlimited access or automatic monitoring.
- Current price
- Confirm in Playground/docs
- Request & concurrency limits
- Confirm in Playground/docs
- Data retention
- Confirm in Playground/docs
- Last verified
- Confirm in Playground/docs
FAQ
Firecrawl API FAQ
What Is the Firecrawl API?
Firecrawl is a web data API that turns web pages into clean Markdown and structured data for applications and AI workflows. It supports finding pages, processing known URLs, discovering site URLs, and collecting content across a site.
What Is the Difference Between Firecrawl Scrape and Crawl?
Scrape processes one known URL. Crawl starts from a site URL, discovers multiple reachable pages, and processes their content. Use Scrape for a specific page and Crawl when you need content from multiple pages.
How Does the Firecrawl Search API Work?
Search finds relevant pages from a query and returns results such as URLs, titles, and descriptions. It can also collect page content when supported by the current endpoint and requested options. Exact fields and formats depend on the current schema.
What Does the Firecrawl Map API Return?
Map discovers URLs associated with a website and mainly returns links, rather than full page content. Use it to inspect site structure or choose pages to scrape. Exact link fields depend on the current endpoint schema.
When Should I Use Firecrawl Batch Scrape?
Use Batch Scrape when you already have a known list of URLs and want to process their content together. Unlike Crawl, it starts from your supplied URL list rather than discovering pages across a site.
What Output Formats Does Firecrawl Support?
Common outputs include clean Markdown, HTML, links, and structured JSON. Available formats, extraction options, and response fields depend on the current endpoint and schema; not every capability returns every format.
Can Firecrawl Handle JavaScript-Rendered Websites?
Firecrawl can handle many JavaScript-rendered pages, but access controls, authentication, robots rules, anti-automation measures, and site changes can affect results. It does not guarantee access to every page; verify the returned content for your target site.
Can I Use Firecrawl for RAG and AI Agents?
Yes. Firecrawl data can feed retrieval-augmented generation (RAG), research agents, and knowledge workflows. Developers remain responsible for source verification, chunking, indexing, and workflow design.
What Is the Difference Between Firecrawl Search, Map, and Crawl?
Search finds pages from a query. Map discovers URLs associated with a site and mainly returns links. Crawl discovers and processes multiple reachable pages to collect their content. Choose based on whether you need source discovery, a URL inventory, or page content.