Beginner AI Permissions and Questions

The bot may ask permission before doing sensitive actions (file edits, command execution, or risky operations).

Two ways prompts appear

1. In IM chat (default and always available).
2. In optional in-world popup dialogs via the LSL dialog bridge.

The dialog bridge is delivered via the bot attachment The Cube Bot (from inventory folder Cube Bot IAR) and can show multiple-choice prompts using llDialog.

If needed, run *bridge install to wear Full Body Alpha and attach The Cube Bot.

Essential star commands

List pending permission requests:

*permission list

Allow a request:

*permission allow <permission-id>

Deny a request:

*permission deny <permission-id>

List pending questions:

*question list

Answer a question:

*question answer <question-id> <your answer>

Reject a question:

*question reject <question-id>

Prompt visibility commands

*prompt status
*prompt show effective
*prompt reload-project

Safety defaults to know

• Permission prompts are interactive by design.
• Handler restrictions can limit who may control the bot.
• Dialog bridge trust checks can require known sender object/owner IDs.

If you are stuck

Run *status and then *permission list to see if the bot is waiting for your approval.