Search to Learn
OPTA KNOWLEDGE BASE
What Is the Opta Stack?
FOUNDATIONS
One stack, three layers — the apps you use, the surfaces you visit, and the engine underneath.
Core outline
- The four primary apps
- Public surfaces vs the engine room
- How the layers talk
Written for
Anyone meeting Opta for the first time
Opta HQ: Your Module OS
APPS
The command centre — modules for work, money, and system control in one cockpit.
Core outline
- Modules and the cockpit home
- Controlling the Opta system
- When HQ is the right door
Written for
Operators who want one place to run everything
Opta Deploy: The AI Workspace
APPS
Long sessions, projects, agent dispatch, and Studios — the desktop home for serious AI work.
Core outline
- Sessions and projects
- Dispatching agents
- Studios and CEO Mode
Written for
Builders running long-form AI sessions
Opta Terminal: Keyboard-Native AI
APPS
The full stack from the command line — chat, runs, and automation without leaving the keyboard.
Core outline
- Why a terminal app at all
- Runs, sessions, and automation
- Terminal vs Deploy
Written for
Keyboard-first developers and tinkerers
Opta Gateway: The Front Door
APPS
Install, onboard, update, and keep your local stack healthy — from first download onward.
Core outline
- First install, step by step
- Updates without fear
- Health checks
Written for
New users setting up their first stack
One Identity: Opta Accounts
PLATFORM
One sign-in, billing, and device trust across every Opta app and surface.
Core outline
- One account, every app
- Device trust explained
- Where billing lives
Written for
Everyone with an Opta account
Local-First AI, Explained
CONCEPTS
Why an AI can run on your own hardware — and what changes when it never leaves your machine.
Core outline
- What "local" actually means
- Privacy and speed trade-offs
- When the cloud still helps
Written for
Anyone curious why local matters
How AI Generates Images
AI BASICS
From pure noise to a finished picture — diffusion explained without the maths degree.
Core outline
- Noise, denoising, and steps
- What prompts actually steer
- Why hands used to go wrong
Written for
Curious minds, no background needed
RAG: Memory for AI
AI BASICS
How retrieval gives an AI grounded, source-traceable memory instead of confident guessing.
Core outline
- Why models forget
- Retrieval in one diagram
- Grounding and citations
Written for
Anyone who has caught an AI making things up
Plugins vs Skills vs MCP
AI TOOLING
Four ways to extend an AI agent — and how to pick the right one for the job.
Core outline
- The four extension shapes
- When each one fits
- Mixing them safely
Written for
Builders extending AI agents
Hooks: Automating an Agent
AI TOOLING
Lifecycle hooks that make a coding agent dependable — guardrails that fire every time.
Core outline
- What a hook is
- Common hook points
- Guardrails vs habits
Written for
Developers running AI coding agents
Power-User AI Capabilities
AI TOOLING
What a fully configured AI coding agent can actually do — beyond the chat box.
Core outline
- Past the chat box
- Agents, workflows, fleets
- Where the ceiling is today
Written for
Users ready to go beyond chatting
Internal vs Product
ARCHITECTURE
Why the engine room — runtime, routing, evidence — never becomes product sprawl.
Core outline
- The boundary that keeps Opta small
- What stays internal and why
- How features cross the line
Written for
Architecture-curious users
Shipping Safely with Git
ENGINEERING
Trunk-based flow, hooks, and autonomous merges that stay safe — even with AI in the loop.
Core outline
- Trunk and short-lived branches
- Hooks as the safety floor
- Letting agents merge
Written for
Teams mixing AI agents with real repos
An AI Voice Receptionist
USE CASES
Building an AI that answers the phone for a real business — and where the hard parts hide.
Core outline
- The pipeline: ears, brain, voice
- Latency is the product
- Handing off to humans
Written for
Builders eyeing real-world AI services