ai intermediate

Jido.AI Task Execution Workflow

Deterministic task lifecycle demo using the shipped task-list actions and real `Jido.Exec.run/3` calls.

primary showcase ai l2 ai-tool-use tasks workflow jido_ai

Related guides and notebooks

What you’ll learn

  • How the shipped task list tools enforce a deterministic lifecycle (pending -> in_progress -> done)
  • How to drive tasklist_add_tasks, tasklist_next_task, tasklist_start_task, tasklist_complete_task, and tasklist_get_state
  • How to keep a workflow demo truthful with real Jido.Exec.run/3 calls and no external services

How this demo works

This page runs real Jido.Exec.run/3 calls against the shipped task-list actions on every button press.

  • Seed Tasks creates a fixed three-step release workflow in local memory.
  • Start Next Task advances the next pending task into in_progress.
  • Complete Active Task records a deterministic result and moves the task into done.
  • Run Full Workflow repeats those transitions until tasklist_get_state reports all_complete.

No external providers, API keys, or network access are required for this demo.

Pull the pattern into your own app

Keep the same Jido.Exec.run/3 shape and swap the seeded tasks or surrounding UI to match your workflow.

  • Reuse the same action sequence when you want deterministic task orchestration in a demo or internal tool.
  • Replace the fixed seed tasks with your own project planning or release-management tasks.
  • Persist the task list in your own agent or workflow state once you move beyond the local demo surface.