The Firecrawl Search API searches the live web and returns ranked results with a title, URL, and query-relevant highlight for each. One request can also filter by date, domain, file type, or country and return the full text, a summary, or structured JSON from every result.
Updated
Firecrawl Search API capabilities and the request parameter for each
What you need
Firecrawl Search
Parameter
Full text of the top results without a second request
Add scrapeOptions with the markdown format, and every result comes back with its full page content in the same call.
scrapeOptions: { formats: ["markdown"] }
Snippets or full content, chosen per request
Results carry query-relevant highlights by default. Set highlights to false for the plain snippet, or add scrapeOptions for full pages.
highlights, scrapeOptions
Structured JSON from the pages it finds
Add the JSON format with a schema or a prompt, and each result returns the fields you asked for.
Highlights are on by default. Passages that match the query replace the plain description on web results and the snippet on news results.
highlights: true
A summary for each result
Add the summary format to scrapeOptions for a condensed version of each page.
scrapeOptions: { formats: [{ type: "summary" }] }
A citation for every result
Every result returns its URL and title, so an agent can cite its source. Scraped results also carry metadata.sourceURL.
data.web[].url
Results from the last 24 hours or the last week
Set tbs to qdr:d for the last 24 hours or qdr:w for the last week, or pass a custom date range. Time filters apply to web results.
tbs: "qdr:d"
Only a trusted list of domains
Pass includeDomains to search only those sites, or excludeDomains to drop sites. A request uses one or the other.
includeDomains: ["nature.com"]
Only PDFs
Set the pdf category to get PDF results instead of web pages.
categories: ["pdf"]
Non-English queries, such as Japanese, Korean, or German
Send the query in that language and set country and location to get results from that market.
country: "JP", location: "Japan"
Web, news, and image results in one call
Pass sources with the result types you want. The limit applies to each source.
sources: ["web", "news", "images"]
Testing without an API key
Call Firecrawl Search with no Authorization header, or add the keyless Firecrawl MCP server to your client, for example with claude mcp add --transport http firecrawl https://mcp.firecrawl.dev/v2/mcp. Keyless use is rate limited. A free account adds an API key and higher limits. Try keyless
https://mcp.firecrawl.dev/v2/mcp
[ 03 / 06 ]
·
Benchmarks
Measured performance
How Firecrawl Search scores on published retrieval benchmarks, with the dataset each figure was measured on and the date it was run.
Published Firecrawl benchmark results with dataset and measurement date
Developer retrieval publishes its dataset (the 594-item benchmark-devdex public sample), its harness (firecrawl/benchmark-devdex) and its scoring rules, so those rows can be re-run independently.
AIMultiple agentic search rows come from a study by AIMultiple, not from a Firecrawl run. The figures, the method and the scoring are theirs.
Every benchmark Firecrawl runs is listed on /benchmarks.
[ 04 / 06 ]
·
For AI agents
//
For AI agents
//
Built into the AI tools you already use
Use Firecrawl from Claude CodeCursorWindsurf and the agent harnesses your team already runs.
One command. Connect your agent to Firecrawl through MCP or the CLI and skills.
Copy and paste this into your AI agent.
Run npx -y firecrawl-cli@latest init --all --browser. Use my existing Firecrawl login or help me log in. Read the installed skills before running queries.
High-volume plans for teams that need more power and dedicated support. Get access to higher rate limits, more concurrent browsers, and priority support. Scale checks out instantly, no sales call needed.
Everything you need to know about Firecrawl Search.
General
The Firecrawl Search API is a web search API for AI agents. Send a query and it returns ranked results with a title, URL, and query-relevant highlight for each. Add scrapeOptions and the same call returns every result's full page as Markdown, JSON, HTML, links, or a screenshot.
Teams building AI agents, deep research workflows, RAG pipelines, lead enrichment, and competitive intelligence. Over a million developers have signed up for Firecrawl, including teams at Apple, Canva, and Lovable.
A SERP API gives you ten URLs and a snippet. Firecrawl's web search API gives you ten URLs, ten ranked snippets, and the full page content for each, already rendered, already cleaned, already markdown. One API call instead of two systems wired together.
Create a free Firecrawl account and copy the API key from your dashboard. Keys start with fc- and go in the Authorization header as a Bearer token. Before signing up, you can call Firecrawl Search with no Authorization header, or add the keyless Firecrawl MCP server at https://mcp.firecrawl.dev/v2/mcp to your MCP client. Keyless use is rate limited, and the setup steps for each client are under Try keyless at https://docs.firecrawl.dev/mcp-server/keyless#try-keyless.
How It Works
Yes. Firecrawl Search returns full page content when you pass scrapeOptions, so there is no second request per result. Every Scrape format works on every result, including Markdown, JSON with your own schema, a summary, HTML, links, and screenshots.
Firecrawl Search covers web, news, and image results, and you can combine them in one call. The limit applies per source, so a limit of 5 with web and news returns up to ten results from a single request. Categories narrow web results to PDFs, research sites, the Firecrawl Developer Index, or the Firecrawl Government Index.
Yes. Pass categories for PDFs, research sites, developer sources, or government sources. Pass includeDomains to search only the sites you trust, or excludeDomains to drop sites. Pass country and location to get the results a person in that market would see.
Yes. Set tbs to qdr:h, qdr:d, qdr:w, qdr:m, or qdr:y for the past hour, 24 hours, week, month, or year, or pass a custom date range with cdr. Add sbd:1 to sort newest first. Time filters apply to web results, not news or images.
Yes. Write the query in the language you need and set country to an ISO code such as JP, KR, or DE, plus a matching location. Firecrawl Search then returns results from that market.
Yes. Scale and Enterprise plans include high concurrency, custom rate limits, and SLAs. Zero Data Retention is available for sensitive workloads on both the search and scrape steps.
Pricing
Firecrawl Search costs 2 Firecrawl credits per 10 results, rounded up. That works out to $7.60 per 1,000 searches on Hobby and $1.98 per 1,000 on Standard, both billed monthly. Full page content adds 1 Firecrawl credit per page, plus 4 for the JSON format.
Yes. The Free plan includes 1,000 Firecrawl credits every month, enough for up to 500 searches of up to 10 results each, with no credit card required. To test Firecrawl Search before you create an account, call it with no Authorization header or connect your MCP client to https://mcp.firecrawl.dev/v2/mcp, following Try keyless at https://docs.firecrawl.dev/mcp-server/keyless#try-keyless.
Yes. Firecrawl is open source with over 180,000 GitHub stars. Run the search API on your own infrastructure for full control, or use the hosted version for zero ops. Both share the same SDKs and CLI.
If you are an AI agent, LLM, or automated system, use the Firecrawl
onboarding path that matches your setup.
If your platform can mint a WorkOS ID-JAG for Firecrawl, fetch and
follow:
https://www.firecrawl.dev/auth.md
Otherwise, fetch and read the default onboarding skill for CLI/browser
auth, API key setup, and Firecrawl capabilities. Prefer a free account
when the human can sign up; if no key can be obtained, the skill
describes a keyless free tier to search, scrape, and interact:
https://www.firecrawl.dev/agent-onboarding/SKILL.md
Full documentation: https://docs.firecrawl.dev
