# Browseai

Browse.ai allows you to turn any website into an API using its advanced web automation and data extraction tools, enabling easy monitoring and data retrieval from websites.

- **Category:** ai web scraping
- **Auth:** API_KEY
- **Composio Managed App Available?** N/A
- **Tools:** 11
- **Triggers:** 0
- **Slug:** `BROWSEAI`
- **Version:** 20260223_00

## Tools

### Bulk Run Tasks

**Slug:** `BROWSEAI_BULK_RUN_TASKS`

This action allows users to bulk run up to 1,000 tasks per API call using a specified robot. For larger datasets, submit multiple bulk runs sequentially (up to 500,000 tasks total). It provides a POST endpoint at /v2/robots/{robotId}/bulk-runs and supports parameters such as robot_id (required), title (required), and input_parameters (required). This bulk operation is essential for large-scale data extraction.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `title` | string | Yes | A descriptive title for the bulk run (1-200 characters). |
| `robot_id` | string | Yes | The unique ID of the robot to run bulk tasks with. |
| `input_parameters` | array | Yes | Array of input parameters for each task to run. Each item is a dictionary of parameters for a single task. Maximum 1,000 tasks per API call (up to 500,000 total via multiple bulk runs). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Monitor

**Slug:** `BROWSEAI_CREATE_MONITOR`

Creates a new monitor for a Browse AI robot to automatically track website changes over time. A monitor runs a robot on a recurring schedule and can notify you when the captured data (screenshots or text) changes. This is useful for tracking price changes, content updates, availability status, or any other dynamic website information. You must first have a robot created (via the Browse AI dashboard) before you can monitor it. Use the GET_ROBOTS action to find available robot IDs. Key features: - Configure recurring schedules (hourly, daily, weekly) using RRULE format - Get email notifications when screenshots or text content changes - Customize notification sensitivity with threshold settings - Override robot parameters for each monitor

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `name` | string | Yes | Monitor name [1-200 characters] |
| `robot_id` | string | Yes | Unique robot ID. You can find a robot's ID by opening it on the dashboard. |
| `schedule` | string | No | Recurring schedule for when the monitor should run, using RRULE format (RFC 5545). Format: 'FREQ=<frequency>;INTERVAL=<number>;BYWEEKDAY=<days>'. Frequency options: HOURLY, DAILY, WEEKLY. Days: MO,TU,WE,TH,FR,SA,SU. Example: 'FREQ=HOURLY;INTERVAL=2' runs every 2 hours, 'FREQ=DAILY;INTERVAL=1;BYWEEKDAY=MO,WE,FR' runs daily on Mon/Wed/Fri. If not provided, monitor must be triggered manually. |
| `input_parameters` | object | No | Input parameters specific to the robot's task configuration. These parameters override the robot's default settings and can include fields like 'originUrl' (starting URL), pagination parameters (skip/limit), or any custom parameters your robot expects. Each robot may require different parameters based on its configuration. |
| `notify_on_text_change` | boolean | No | Whether to send email notifications when the captured text content changes between monitoring runs |
| `notify_on_screenshot_change` | boolean | No | Whether to send email notifications when the captured screenshot changes between monitoring runs |
| `screenshot_notification_threshold` | integer | No | Minimum percentage of screenshot pixels that must change to trigger a notification (0-100). For example, 15 means notify only if at least 15% of the screenshot has changed |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Webhook

**Slug:** `BROWSEAI_CREATE_WEBHOOK`

This tool creates a new webhook for a Browse AI robot. Webhooks are HTTP callbacks that Browse AI sends to your server immediately when specific events occur, eliminating the need for polling. It is useful for: - Setting up automated notifications for task completion - Receiving real-time updates when changes are detected - Integrating Browse AI with your own systems - Automating workflows based on robot task results The webhook can be configured to trigger on different event types: - taskFinished: Triggers on any task completion (success or failure) - taskFinishedSuccessfully: Triggers only on successful task completions - taskFinishedWithError: Triggers only on failed task completions - taskCapturedDataChanged: Triggers when data changes are detected during monitoring - tableExportFinishedSuccessfully: Triggers when table export completes (Beta feature) Note: Browse AI only supports one event type per webhook. To monitor multiple event types, create separate webhooks for each event type.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `url` | string | Yes | The URL where webhook notifications will be sent |
| `events` | array | No | List of events to trigger the webhook. Note: Browse AI only supports one event type per webhook, so only the first event will be used. Valid event types: taskFinished, taskFinishedSuccessfully, taskFinishedWithError, taskCapturedDataChanged, tableExportFinishedSuccessfully |
| `robot_id` | string | Yes | The ID of the robot to attach the webhook to |
| `is_active` | boolean | No | Whether the webhook should be active immediately |
| `description` | string | No | A description of the webhook |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete a specific monitor

**Slug:** `BROWSEAI_DELETE_MONITOR`

This tool allows users to delete a specific monitor from their Browse AI account. It uses the DELETE method and requires a valid monitor_id.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `monitor_id` | string | Yes | The unique ID of the monitor to be deleted |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete a specific task

**Slug:** `BROWSEAI_DELETE_TASK`

This tool allows you to delete a specific task in BrowseAI by its task ID. It is used for cleaning up completed or failed tasks, managing resources, and maintaining your task list.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `task_id` | string | Yes | The unique ID of the task to be deleted |
| `robot_id` | string | Yes | The unique ID of the robot that created the task |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Robots List

**Slug:** `BROWSEAI_GET_ROBOTS`

Retrieves a list of all robots (automated web tasks) under your Browse AI account. A robot is an automated browser task that can be trained to perform web operations such as: - Opening webpages and navigating through sites - Logging into websites - Clicking buttons and filling forms - Extracting structured data from web pages - Monitoring websites for changes This action returns comprehensive information about each robot including: - Robot ID (required for running tasks via other API endpoints) - Robot name and timestamps (creation/update) - Input parameters the robot accepts (parameter names, types, and whether they're required) Use this to: - Discover available robots and their IDs for task execution - Check robot configuration and required input parameters - Monitor robot creation and update activity - Get an overview of all automated tasks in your account No parameters required - simply call this action to retrieve all robots.

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Robot Tasks

**Slug:** `BROWSEAI_GET_ROBOT_TASKS`

Retrieves a paginated list of all tasks (execution runs) for a specific Browse AI robot. Each task represents one execution of the robot, which includes its current status, execution timestamps, input parameters, captured text data, screenshots, and extracted lists. This tool returns tasks in reverse chronological order (newest first) by default. Use the limit and offset parameters to paginate through large result sets. Use this tool to: - Check the status of recent robot executions (pending, running, successful, failed) - Access data captured by robot tasks (text, screenshots, lists) - Monitor robot performance and execution history - Retrieve task IDs for use with other task-related actions - Debug failed robot executions Note: You must have at least one robot configured in your Browse AI account to use this action. Use the Get Robots List action first to find available robot IDs.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `limit` | integer | No | Maximum number of tasks to return in a single request. Use this with offset for pagination. |
| `offset` | integer | No | Number of tasks to skip before starting to return results. Use this with limit for pagination (e.g., offset=10, limit=10 returns tasks 11-20). |
| `robot_id` | string | Yes | The unique identifier (UUID format) of the robot whose tasks you want to retrieve. You can get robot IDs from the Get Robots List action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get System Status

**Slug:** `BROWSEAI_GET_SYSTEM_STATUS`

Tool to check the operational status of Browse AI infrastructure, including the tasks queue condition. Use when you need to verify if Browse AI services are operational before running robots or tasks.

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Task Details

**Slug:** `BROWSEAI_GET_TASK_DETAILS`

Retrieves detailed information about a specific Browse AI task by robot ID and task ID. This tool returns comprehensive task details including execution status, captured data, screenshots, input parameters, timestamps (created, started, completed), and error information if the task failed. It provides an in-depth view of individual task execution for monitoring, debugging, and data retrieval purposes. Use this tool when you need to: - Check the current status of a task (running, successful, failed) - Retrieve data captured during task execution - Access screenshots taken during the task - Review input parameters used for the task - Debug task failures by examining error details - Monitor task execution timestamps

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `task_id` | string | Yes | The unique identifier of the task to retrieve details for |
| `robot_id` | string | Yes | The unique identifier of the robot that executed the task |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Robot Webhooks

**Slug:** `BROWSEAI_LIST_ROBOT_WEBHOOKS`

Tool to retrieve all webhooks configured for a specific Browse AI robot. Use when you need to view, audit, or manage webhook configurations for a robot. Returns webhook details including URL, event type, active status, and creation timestamp.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `robot_id` | string | Yes | The unique identifier (UUID format) of the robot whose webhooks you want to retrieve. You can get robot IDs from the Get Robots List action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Run Robot

**Slug:** `BROWSEAI_RUN_ROBOT`

Triggers the execution of a Browse.ai robot to extract data from websites. This action starts a new robot task that will scrape/extract data according to the robot's pre-configured settings. The robot runs asynchronously and returns a task ID immediately. To retrieve the extracted data, use BROWSEAI_GET_TASK_DETAILS with the returned task ID. Prerequisites: - You must have at least one robot configured in your Browse.ai account - Obtain robot_id using BROWSEAI_GET_ROBOTS action first - Know what input parameters your robot expects (configured during robot creation) Common use cases: - Scraping product data from e-commerce sites - Monitoring website changes for specific content - Extracting structured data from multiple web pages - Automating repetitive web data collection tasks Note: The robot executes asynchronously. Check task status using BROWSEAI_GET_TASK_DETAILS or BROWSEAI_GET_ROBOT_TASKS to retrieve results when the task completes.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `robot_id` | string | Yes | The unique identifier of the Browse.ai robot to execute. You can retrieve robot IDs using the BROWSEAI_GET_ROBOTS action. |
| `input_parameters` | object | No | Optional dictionary of custom input parameters for the robot execution. Common parameters include 'originUrl' for the starting URL, or custom parameters defined when the robot was created (e.g., search terms, filters, pagination settings). The required parameters depend on how the specific robot was configured. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |