Search

Search Google programmatically and optionally scrape each result page.

Endpoint

POST /v1/search

curl -X POST https://api.browsr.dev/v1/search \
  -H "Authorization: Bearer $BROWSR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "browsr headless browser API",
    "limit": 5,
    "formats": ["markdown"]
  }'

Request

Field	Type	Default	Description
query	string	required	Search query string
limit	number	5	Max results (max: 20)
formats	string[]	[]	If non-empty, each result URL is scraped in these formats
session_id	string	—	Reuse an existing session

When formats is empty, only search metadata is returned (no page scraping).

Response

{
  "success": true,
  "data": [
    {
      "url": "https://browsr.dev",
      "title": "Browsr - Headless Browser API",
      "description": "Browser automation platform built in Rust...",
      "position": 1,
      "content": {
        "markdown": "# Browsr\n\n...",
        "metadata": {
          "title": "Browsr",
          "sourceURL": "https://browsr.dev",
          "status_code": 200
        }
      }
    }
  ]
}

The content field is only present when formats is non-empty. It contains the scraped data for each result, using the same format as the Scrape response.

Examples

Search only (no scraping)

curl -X POST https://api.browsr.dev/v1/search \
  -H "Authorization: Bearer $BROWSR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "rust web scraping", "limit": 10}'

Search and scrape results

curl -X POST https://api.browsr.dev/v1/search \
  -H "Authorization: Bearer $BROWSR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "best headless browser APIs 2025",
    "limit": 5,
    "formats": ["markdown", "links"]
  }'