Quick Start (Startup)

April 8, 2026 · View on GitHub

If this is your first time using OpenASE, follow these 5 steps to get the full workflow running.

Step 1: Create a Project & Configure Basic Settings

Go to Settings and complete the following:

  • Fill in the project name and description
  • Configure ticket statuses (e.g. To Do, In Progress, Done) — these become columns on the board
  • Connect a code repository so agents can access code. OpenASE supports hosted providers such as GitHub / GitLab as well as machine-local or self-hosted Git sources via file://.

Step 2: Register a Machine

Go to Machines and add an execution environment for agents:

  • Can be the reserved local machine, a direct-connect listener, or a reverse-connect daemon host
  • Remote machines should use websocket runtime; keep SSH only if you want helper bootstrap or diagnostics access
  • Click "Test Connection" after adding to verify connectivity

Step 3: Register an Agent

Go to Agents and register an agent from an available AI Provider (e.g. Claude Code, Codex, Gemini CLI).

Step 4: Create a Workflow

Go to Workflows and create a workflow template:

  • Select the agent to bind
  • Write a Harness (instruction document) telling the agent how to execute tasks
  • Configure which ticket statuses trigger automatic agent pickup

Step 5: Create a Ticket

Go to Tickets and create your first ticket:

  • Fill in the title and description
  • Select the workflow you just created
  • Set the ticket status to "To Do"

The agent will automatically detect this ticket, pick it up, and begin execution. You can watch the process in real-time in Activity.

How It Works

User creates ticket  →  Orchestrator detects pickup status
    →  Agent claims ticket  →  Executes workflow on Machine
    →  Process recorded in Activity  →  Ticket completes

Next Steps

After completing the flow above, you've experienced OpenASE's core capabilities. Next you can:

  • Dive deeper into Workflows Harness writing to improve agent execution quality
  • Explore Skills to add reusable capabilities to workflows
  • Set up Scheduled Jobs to fully automate repetitive work
  • Use Updates to record project progress for team collaboration