22 lines
1.5 KiB
Django/Jinja
22 lines
1.5 KiB
Django/Jinja
The user is trying to achieve a goal in the browser assisted by an browser AI assistant.
|
|
|
|
According to the AI assistant's feedback, including reasoning and its message, there's no immediate action the assistant can take in the website.
|
|
|
|
Help the user decide what to do next based on the assistant's message. Here's the list of available actions:
|
|
- solve_captcha: the task is blocked by captcha and the assistant is asking the user to solve the captcha
|
|
- complete: the user goal has been achieved
|
|
- terminate: the user goal cannot be achieved. Terminate the task. Examples: 1) there's not enough data provided to achieve the goal and the assistant is asking the user to provide more information. For examples: login is required and the user has not provided the login credentials or incorrect credentials are provided; a form needs to be filled and a required field is missing. 2) The site is stuck or not loading after multiple attempts
|
|
- get_verification_code: the assistant is asking the user to provide a verification code (2FA, MFA or TOTP code)
|
|
- other: the assistant is asking the user to do something else
|
|
|
|
Return the action to take next in the following JSON format:
|
|
{
|
|
"action": str // complete, terminate, solve_captcha, get_verification_code
|
|
"useful_information": str // If there is any useful information the assistant has provided that contributes to the user goal, put it here.
|
|
}
|
|
|
|
User goal: {{ navigation_goal }}
|
|
|
|
Assistant reasoning: {{ assistant_reasoning }}
|
|
|
|
Assistant message: {{ assistant_message }} |