Search API for
AI agents & apps

# Get a free API key (one command)
curl -X POST https://searchx.dev/api/v1/signup \
  -d '{"email":"you@example.com"}'
# → {"api_key": "sk-sx-...", "daily_limit": 3000}

# Search the web
curl https://searchx.dev/api/v1/search \
  -H "Authorization: Bearer sk-sx-YOUR_KEY" \
  -G -d "q=kubernetes deployment" -d "mode=hybrid"

# Read a page in four shapes, one fetch
curl https://searchx.dev/api/v1/extract \
  -G -d "url=https://react.dev" \
     -d "formats=markdown,links,html,screenshot"
# → every key you asked for. One that could not be produced
#   comes back null with the reason in "warnings".

# Re-crawl a site, get back only what changed
curl -X POST https://searchx.dev/api/v1/crawl \
  -H "Authorization: Bearer sk-sx-YOUR_KEY" \
  -d '{"url":"https://docs.example.com","track_changes":true,"only_changed":true}'

pip install searchx

from searchx import SearchX

sx = SearchX.from_free_signup("you@example.com")

# Web search (hybrid: keyword + semantic)
results = sx.search("kubernetes deployment", mode="hybrid")
for r in results:
    print(r.title, r.url)

# Extract page content (auto JS rendering)
page = sx.extract("https://react.dev")

# AI answer with citations
answer = sx.answer("what is docker", include_answer=True)

# Image search
images = sx.images("sunset mountain")

npm install searchx

import { SearchX } from 'searchx'

const sx = await SearchX.fromFreeSignup('you@example.com')

// Web search
const results = await sx.search('kubernetes deployment')

// Extract page content (auto JS rendering)
const page = await sx.extract('https://react.dev')

// AI answer
const answer = await sx.answer('what is docker')

// Image search
const images = await sx.images('sunset mountain')

// Claude Desktop / Cursor config
// ~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "searchx": {
      "url": "https://mcp.searchx.dev/mcp",
      "transport": "http"
    }
  }
}

Tools: search, answer, extract, suggest