Workflows

import { Aside } from ‘@astrojs/starlight/components’;

Workflows define how tasks move through your development process and where AI automation kicks in.

Workflow structure

A workflow consists of:

• Statuses — stages a task passes through (e.g., Backlog, In Progress, Done)
• Transitions — allowed moves between statuses
• Bindings — AI job templates attached to specific transitions

Built-in templates

Software Development

A 7-status workflow for feature development:

Backlog → Ready → In Progress → Code Review → QA → Staging → Done

AI bindings included:

• In Progress → runs the coding agent to implement the task
• Code Review → runs automated code review against project rules
• QA → runs test suite and generates test report

Kanban

A minimal 3-status workflow:

To Do → In Progress → Done

No AI bindings by default — add them as needed.

AI Job Bindings

Bindings connect an AI Job Template to a workflow transition. When a task moves to a bound status, the job runs automatically.

Task moves to "Code Review"
    │
    ▼
Binding fires: code-review-template
    │
    ▼
AI Job created (pending → running)
    │
    ▼
Agent analyzes code, writes review artifact
    │
    ▼
Artifact appears in task's Agent tab
    │
    ▼
Developer approves → task auto-moves to "QA"

You can configure a binding to **auto-transition** the task when the artifact is approved. This enables fully automated pipelines where human review is the only manual step.

Custom workflows

To create a custom workflow:

1. Go to Settings → Workflows → New Workflow
2. Add statuses and define allowed transitions
3. Bind AI job templates to transitions
4. Assign the workflow to a project

Workflows can be shared across multiple projects within your organization.