Basic Browser Agent
AI & KnowledgeAI-powered browser automation for simple public web tasks
Overview
The Basic Browser Agent uses a secured Playwright session and a ReAct loop to browse public pages, click elements, fill forms, and return a final answer. It is the best starting point for short tasks on public websites without vision or macro guidance.
Configuration
| Property | Type | Default | Description |
|---|---|---|---|
goalrequiredvars | multiline | — | What the agent should accomplish. |
startUrlvars | string | — | Optional starting URL. |
model | options | — | LLM used to choose browser actions. |
maxSteps | number | 15 | Maximum browser actions. Legacy graphs may still use maxIterations. |
timeout | number | 30000 | Per-action timeout in milliseconds. |
headless | boolean | true | Run Chromium without a visible window. |
screenshots | boolean | false | Capture a final page screenshot when enabled. |
Output Variables
Reference these variables in downstream nodes using the {{node.field}} syntax.
{{node-id.result}}stringFinal answer from the agent
{{node-id.goal}}stringResolved browsing goal
{{node-id.screenshots}}arrayCaptured screenshots
{{node-id.iterations}}numberSteps taken during the run
{{node-id.usage.browser_ms}}numberBrowser runtime in milliseconds
Examples
Pricing lookup
Start on a public pricing page and extract the monthly plan cost.
{
"goal": "Find the monthly price for the Pro plan",
"startUrl": "https://example.com/pricing",
"maxSteps": 10,
"screenshots": true
}Tips & Best Practices
- •Use the Test tab to preview result text, browser steps, and usage before publishing.
- •Escalate to Vision Browser Agent when the page is canvas-heavy or needs OCR.
- •Escalate to Guided Browser Agent when assertions or macro hints are required.
Related Nodes
Try Basic Browser Agent in Your Workflow
Create a free account and start building AI-powered workflows with the Basic Browser Agent node.
Open Workflow Editor