Perform actions on a website using an AI Agent
Provide an AI agent with instructions to perform actions in your web browser session. This action is designed only for performing interactions (e.g., clicking, typing, navigating) and not for extracting page content. If no interactions occur, the actions output will be empty.This Wrk Action uses AI to deliver results. Results may vary significantly based on prompts provided and websites accessed.
Application
Inputs (what you have)
| Name | Description | Data Type | Required? | Example |
|---|
| Browser session ID | The unique identifier of an active browser instance. This ID can be retrieved from an "Open a browser session" Wrk Action. Example: fb9bc380-146c-420e-9130-50ce93614e05 | Text (Short) | Yes | |
| Instructions | Detailed instructions for the AI agent to perform on the specified website | Text (Long) | Yes | |
| Model name | The name of the AI model to use for the agent | Text (Short) | No | |
| Max steps | The maximum number of steps the AI agent will attempt to take before it automatically stops execution. Maximum value is 20. | Number | No | |
Outputs (what you get)
| Name | Description | Data Type | Required? | Example |
|---|
| Agent message | Results of how the agent performed against the provided instructions | Text (Long) | No | |
| Actions | A JSON array of the actions the agent took to perform your request | Text (Long) | No | |
| Error message | If unsuccessful, a message stating what went wrong | Text (Long) | No | |
| Unsuccessful screenshot | If unsuccessful, a screenshot of the website | File | No | |
Outcomes
| Name | Description |
|---|
| Success | This status is selected if the job has successfully completed. |
| Unsuccessful | This status is selected if the job has completed unsuccessfully. |
Requirements