Quickstart - Extralt Docs

This guide walks you through your first extraction using the Extralt API. By the end, you'll have structured product data from a live ecommerce site.

Prerequisites

Create an account at extralt.com
Create an organization
Subscribe to a plan (Start includes a 7-day free trial with 5,000 credits)
Generate an API key from the dashboard

See Account Setup for detailed instructions.

Set your API key

export EXTRALT_API_KEY="your-api-key"
export BASE_URL="https://api.extralt.com"

Create a robot

A robot is an AI-generated crawler for a specific site. Provide a URL and country to build one.

curl -s -X POST "$BASE_URL/robot-builds" \
  -H "Authorization: Bearer $EXTRALT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example-store.com/products/sample",
    "country": "US"
  }' | jq

Robot builds take 3-5 minutes. The AI analyzes the site structure and generates a compiled crawler.

Check build status

Poll the build until it completes.

curl -s "$BASE_URL/robot-builds/$BUILD_ID" \
  -H "Authorization: Bearer $EXTRALT_API_KEY" | jq '.status'

When the build completes, a robot is created. The build response includes the robotId.

Create a run

A run extracts data from one or more URLs using a robot.

curl -s -X POST "$BASE_URL/runs" \
  -H "Authorization: Bearer $EXTRALT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "robotId": "'$ROBOT_ID'",
    "urls": ["https://example-store.com/products/sample"],
    "budget": 10
  }' | jq

Fetch captures

Once the run completes, fetch the extracted data.

curl -s "$BASE_URL/captures?runId=$RUN_ID" \
  -H "Authorization: Bearer $EXTRALT_API_KEY" | jq

What's next

Core Concepts -- understand the data model
Running Extractions -- advanced run options
Common Patterns -- polling, pagination, error handling