Core Concepts
Understanding these concepts will help you get the most out of Otto.
Organizations
An Organization is your team's workspace in Otto. It contains:
- Team members with different roles (Owner, Admin, Member)
- Integrations (Linear, GitHub, Azure DevOps, Jira — with more coming soon)
- Rules, lifecycle settings, and projects
- Billing and usage tracking
Member Roles
| Role | Permissions |
|---|---|
| Owner | Full access, billing, delete org |
| Admin | Manage rules, integrations, settings |
| Member | View-only access |
Lifecycle
Otto uses a built-in 3-phase lifecycle to work on tickets. Each phase is handled by a preset agent:
| Phase | What It Does |
|---|---|
| Clarify | Reads the ticket, explores the codebase, asks clarifying questions |
| Plan | Creates a detailed implementation plan and posts it for approval |
| Implement | Writes code, creates a branch, opens a PR, and handles review feedback |
Lifecycle Settings
You can customize the lifecycle in Settings → Lifecycle:
- Enable/disable phases — Disabling an earlier phase automatically disables subsequent ones
- Choose AI models — Select which model to use for each phase (defaults to Claude Sonnet)
Rules
Rules define which tickets Otto should work on. They act as an opt-in filter for your PM tool (Linear or Jira).
- If no rules exist, Otto won't pick up any tickets — rules are required
- Multiple rules use OR logic — a ticket matching any rule will be picked up
- Rules can filter by team, labels, statuses, and assignment
Rules are configured per integration in Settings → Integrations → Linear/Jira.
Projects
A Project links your ticket management tool (Linear or Jira) to your code hosting tool (GitHub or Azure DevOps). This tells Otto:
- Which repo to work in for tickets from a specific project
- Where to create pull requests
- Which branches to work with
Context
Context is additional information you provide to help Otto understand your codebase and conventions:
- Coding standards and patterns
- Architecture decisions
- Technology stack details
- Team preferences
Good context leads to better code generation.
Agents
Otto's core lifecycle uses preset agents that work out of the box. For one-off workflows outside the standard lifecycle, you can create custom Agents with their own models, system prompts, and configurations.
Tokens & Usage
Otto uses AI models that charge based on tokens (roughly words). Your usage is tracked and billed based on:
- Input tokens (what you send to the AI)
- Output tokens (what the AI generates)
See Billing for more details.