Introducing FireSearch: From SERP to LLM-Ready Data in Seconds

We are thrilled to announce the beta release of FireSearch, a search + extraction API. With /search endpoint, you give Firecrawl a query, it searches the web for the top results, scrape the top pages, and gives you clean markdown for each so you can feed directly to your LLM.

​

Quick Start Guide

Here’s how you can get started with FireSearch to fetch web content for your LLM.

If you’d like to use our SDKs, you can find the Python SDK here and the Node SDK here.

​

Step 1: Crafting Your Search Request

To execute a search query and receive markdown-formatted content:

curl -X POST https://api.firecrawl.dev/v0/search \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer fc-YOUR_API_KEY' \
-d '{
  "query": "your_search_term",
  "pageOptions": {
    "fetchPageContent": true
  }
}'

Replace fc-YOUR_API_KEY with your actual API key and your_search_term with the term you wish to search for.

​

Step 2: Handling the Response

The FireSearch API will return search results in markdown format, which you can then readily supply to your LLM for further processing or integration into your system.

{
  "success": true,
  "data": [
    {
      "url": "https://www.mendable.ai/",
      "markdown": "# Markdown Content of the page",
      "metadata": {
          "title": "Mendable | AI for CX and Sales",
          "description": "AI for CX and Sales",
          "language": null,
          "sourceURL": "https://www.mendable.ai/"
      }
    },
    {
      "url": "https://github.com/mendableai",
      "markdown": "# Markdown Content of the page",
      "metadata": {
          "title": "Mendable | AI for CX and Sales",
          "description": "AI for CX and Sales",
          "language": null,
          "sourceURL": "https://www.mendable.ai/"
      }
    },
  ]
}

If "fetchPageContent": false, data returns faster but without the full page markdown content.

[
  {
    "title": "Example Page Title",
    "url": "https://www.example.com",
    "content": "# Page Heading\n\nContent of the page in markdown format..."
  },
  // Additional results
]

We look forward to seeing how FireSearch can power up your projects!

​

Current Status:

FireSearch is currently in beta, and we are diligently working to refine and improve its capabilities.

Integration support for playgrounds and @LangChainAI is on the horizon.