AI Assistant

Interactive

VoxelAI includes a powerful AI assistant that can help players and staff with various server-related tasks using Google's Gemini AI.

Features

Natural Language Understanding

• Understands player questions and requests

• Interprets context from recent chat

• Handles complex multi-step tasks

• Provides relevant server information

Command Execution

• Can execute commands on behalf of players

• Understands command permissions

• Handles multi-command sequences

• Prevents dangerous operations

Server Information

• Provides real-time server stats

• Checks player information

• Monitors server health

• Tracks plugin status

Usage

Basic Commands

/ai <question>

Ask the AI anything

/ai toggle

Toggle AI assistant on/off

/ai status

Check AI status and conversation history

/ai clear

Clear conversation history

/ai clearcontext

Clear chat context

Configuration

Basic Settings

assistant:
  enabled: true
  max-tokens: 2048
  temperature: 0.7        # 0.0 = deterministic, 1.0 = creative
  command-execution: true # Enable command suggestions

Blocked Commands

Commands that the AI assistant will never suggest

blocked-commands:
  - "stop"
  - "restart"
  - "reload"
  - "ban"
  - "pardon"
  - "op"
  - "deop"
  # ... and more

Permissions

Permission Nodes

voxelai.use:
  description: Access to VoxelAI system
  default: false

voxelai.assistant.use:
  description: Access to AI assistant
  default: false
  children:
    - voxelai.use

voxelai.command.execute:
  description: Allow AI to execute commands
  default: op
  children:
    - voxelai.assistant.use

AI Models

Primary Model

High RPM, high daily limit

gemma-3-27b-it

30 RPM, 14,400 daily requests

Fallback Model

Lower RPM, lower daily limit

gemini-2.5-flash-lite-preview-06-17

15 RPM, 1,000 daily requests

Best Practices

Command Usage

• Ask questions in natural language

• Let AI handle command sequences

• Trust AI's permission checking

• Review suggested commands

Question Asking

• Be specific in requests

• Provide context if needed

• Use natural language

• Ask follow-up questions

Server Management

• Use AI for routine tasks

• Monitor AI command usage

• Set appropriate permissions

• Review blocked commands

Troubleshooting

Rate Limiting

• Watch API usage

• Use hybrid model system

• Monitor daily limits

• Check queue status

Permissions

• Verify player permissions

• Check command access

• Review blocked commands

• Test with different roles

Command Issues

• Check command syntax

• Verify plugin compatibility

• Test manually first

• Review error logs