Task Execution Module

The Task Execution Module connects user goals with web automation, facilitating the interaction between users and the web through agents like BrowserAgent. It translates user tasks (e.g., finding a flight) into actionable commands, automates web interactions, and returns results.

Key Components:

1. User Goals: A clear task, such as finding the cheapest flight, is input into the system.

2. Agent Initialization: The browser agent is configured with the task and an LLM (like GPT-4) to interpret and execute it.

3. Task Execution: The agent navigates the web, interacts with forms or fields, scrapes relevant data, and completes the task.

4. Result Handling: Once the task is complete, results are returned and presented to the user.

5. Error Management: Errors are caught and logged, ensuring the process remains robust.

6. Scalability: The module can be expanded to handle more complex or multi-step tasks, such as booking multiple services.

In essence, the Task Execution Module automates web-based tasks, making it easier for users to achieve goals with minimal intervention.

Task Execution Module in Action

Through a simple interface the browser agent can be utilized by the developer to enable interaction with the underlying Browser Agent to parse the web, gathering data and taking action.

import { BrowserAgent } from '@nasdaiq/browser-agent';

async function main() {
  const agent = new Agent({
    key: process.env.BROWSER_AGENT_KEY,
    task: 'Find a one-way flight from Paris to Tokyo on 25 February 2025 on Google Flights. Return me the cheapest option.',
    llm: 'gpt-4o',
  });

  const result = await agent.run();
    
  console.log(await result.data());
}

main().catch((error) => {
  console.error(error);
});