# Browser Use (/integrations/browser_use)





{/* MANUAL-CONTENT-START:intro */}

[Browser Use](https://browser-use.com/) runs browser tasks from natural-language instructions. Use it to navigate websites, interact with page controls, and extract information as a workflow step. The result includes the task status, output, and a screenshot URL when available.

{/* MANUAL-CONTENT-END */}

## Use the block [#usage-instructions]

Run a browser task from natural-language instructions to navigate pages, interact with controls, and extract information.

## Actions [#actions]

### Browser Use [#browser-use]

Runs a browser automation task using BrowserUse

<ReferenceDetails>
  #### Input [#input]

  | Parameter | Type | Required | Description |
  | - | - | - | - |
  | `task` | string | Yes | What should the browser agent do |
  | `startUrl` | string | No | Initial page URL to start the agent on (reduces navigation steps) |
  | `variables` | json | No | Optional secrets injected into the task (format: \{key: value}) |
  | `allowedDomains` | string | No | Comma-separated list of domains the agent is allowed to visit |
  | `maxSteps` | number | No | Maximum number of steps the agent may take (default 100, max 10000) |
  | `flashMode` | boolean | No | Enable flash mode (faster, less careful navigation) |
  | `thinking` | boolean | No | Enable extended reasoning mode |
  | `vision` | string | No | Vision capability: "true", "false", or "auto" |
  | `systemPromptExtension` | string | No | Optional text appended to the agent system prompt (max 2000 chars) |
  | `structuredOutput` | string | No | Stringified JSON schema for the structured output |
  | `highlightElements` | boolean | No | Highlight interactive elements on the page (default true) |
  | `metadata` | json | No | Custom key-value metadata (up to 10 pairs) for tracking |
  | `model` | string | No | LLM model identifier (e.g. browser-use-2.0) |
  | `apiKey` | string | Yes | API key for BrowserUse API |
  | `profile_id` | string | No | Browser profile ID for persistent sessions (cookies, login state) |

  #### Output [#output]

  | Parameter | Type | Description |
  | - | - | - |
  | `id` | string | Task execution identifier |
  | `success` | boolean | Task completion status |
  | `output` | json | Final task output (string or structured) |
  | `steps` | array | Steps the agent executed (number, memory, nextGoal, url, actions, duration) |
  | ↳ `number` | number | Sequential step number |
  | ↳ `memory` | string | Agent memory at this step |
  | ↳ `evaluationPreviousGoal` | string | Evaluation of previous goal completion |
  | ↳ `nextGoal` | string | Goal for the next step |
  | ↳ `url` | string | Current URL of the browser |
  | ↳ `screenshotUrl` | string | Optional screenshot URL |
  | ↳ `actions` | array | Stringified JSON actions performed |
  | ↳ `duration` | number | Step duration in seconds |
  | `liveUrl` | string | Embeddable live browser session URL (active during execution) |
  | `shareUrl` | string | Public shareable URL for the recorded session (post-run) |
  | `sessionId` | string | Browser Use session identifier |
</ReferenceDetails>