Stagehand Crawler
Pricing
from $20.00 / 1,000 results
Go to Apify Store
Stagehand Crawler
Stagehand is a browser automation framework used to control web browsers with natural language and code.
An Apify Actor that uses Stagehand — an AI-powered browser automation framework — to execute natural language instructions on web pages. Point it at any URL, describe what you want in plain English, and the AI agent will interact with the page and return results along with a final screenshot.
What it does
- Visits each URL in
startUrls - Runs an AI agent that follows your
instructionusing real browser interactions (clicks, typing, scrolling, navigation) - Captures a screenshot of the final page state and saves it to the key-value store
- Pushes a result record to the dataset containing the outcome, a summary message, and the screenshot URL
Input
| Field | Type | Required | Description |
|---|---|---|---|
startUrls | array | yes | URLs the crawler will visit |
instruction | string | yes | Natural language task for the AI agent (e.g. "Find the price of the first product") |
mode | string | yes | Agent mode: dom, hybrid, or cua |
model | string | yes | AI model in provider/model format (e.g. google/gemini-3-flash-preview) |
apiKey | string | yes | API key for the selected model provider |
proxyConfig | object | yes | Proxy settings — Apify Proxy recommended on the platform |
maxSteps | integer | no | Max browser interactions before stopping (default: 20, range: 1–20) |
Agent modes
dom— Reads the page DOM to determine actions. Fast and cost-efficient.hybrid— Combines DOM and visual understanding. Best for complex layouts. Requires a vision-capable model (Gemini, Claude).cua— Computer-use agent; operates purely from screenshots. Most powerful, highest cost.
Example input
Output
Each processed URL produces one dataset record:
The imageUrl links directly to the PNG screenshot saved in the default key-value store.
