Opta Help

Comprehensive documentation for the Opta Local private AI stack. CLI, Daemon, LMX inference, Local Web, and more.

Get StartedCLI Reference

Getting Started

What is the Opta Local stack and who is it for

  • Installation
  • LAN Setup
  • First Session

CLI Reference

CLI commands at a glance

  • Chat & Do
  • Model Management
  • Sessions
  • Configuration
  • Slash Commands

Daemon

What the daemon does and why it exists

  • Lifecycle
  • HTTP API
  • WebSocket Events
  • Troubleshooting

LMX

The local inference engine for Apple Silicon

  • Setup
  • API Reference
  • Model Management
  • Monitoring

Local Web

Browser-based dashboard and chat client

  • Dashboard
  • Chat
  • Remote Access

Code Desktop

Visual session monitor for daemon activity

  • Session Management
  • Daemon Controls

Browser Automation

AI-driven browser control via Playwright

  • Tools
  • Recording & Replay

Security

Security model and design principles

  • Permissions
  • Privacy
  • Guardrails

Developer Guide

Building on the Opta Local stack

  • MCP Integration
  • Daemon Client SDK
  • API Authentication

Feature Status

Current feature implementation status

    The Opta Local Stack

    Three components form a layered local-inference pipeline

    opta chat / opta tui / opta do# Opta CLI
    │
    opta daemon127.0.0.1:9999# Session orchestration
    │ HTTP v3 REST + WebSocket
    Opta LMX192.168.188.11:1234# MLX inference
    │ OpenAI-compatible API
    Opta Local Weblocalhost:3004# Dashboard + Chat