Guided Browser Agent
AI & KnowledgeGoal-driven browsing with macro hints, assertions, and secure credential handling
Overview
The Guided Browser Agent runs a browser-use planner against your goal, optionally guided by saved macro hints, then verifies the final page with typed assertions. Credential placeholders remain redacted in prompts and streamed logs.
Configuration
| Property | Type | Default | Description |
|---|---|---|---|
goalrequiredvars | multiline | — | Natural-language browsing goal. |
startUrlvars | string | — | Optional starting URL. |
macroId | string | — | Optional recorded macro used as structured guidance. |
assertions | json | — | Post-run visible/text/url assertions. |
model | options | — | Planner model. |
maxSteps | number | 15 | Maximum planner steps. |
screenshots | boolean | true | Capture a final screenshot after the AI run. |
headless | boolean | true | Run Chromium headlessly. |
Output Variables
Reference these variables in downstream nodes using the {{node.field}} syntax.
{{node-id.success}}booleanWhether the run and assertions succeeded
{{node-id.result}}stringAI result text
{{node-id.url}}stringFinal page URL
{{node-id.assertions}}arrayAssertion results
{{node-id.macroHintsUsed}}booleanWhether macro hints were attached
Examples
Login flow with assertions
Guide the agent with a macro and verify the dashboard URL.
{
"goal": "Sign in and confirm the dashboard loads",
"macroId": "your-macro-id",
"assertions": [
{ "type": "assert_url", "expected": "/dashboard" }
]
}Tips & Best Practices
- •Use {{secrets.NAME}} in goals only when necessary; values are redacted from prompts as [CREDENTIAL:NAME].
- •Run the Test tab to inspect assertion results before publishing.
- •Use Deterministic Browser Macro when the flow becomes stable and repeatable.
Related Nodes
Try Guided Browser Agent in Your Workflow
Create a free account and start building AI-powered workflows with the Guided Browser Agent node.
Open Workflow Editor