Build an agent or harness

Turn a small task into a bounded, observable model/tool loop.

Questions to answer

What must be persisted? Which errors reach the model? When must execution stop?

Start here

Start with the build path. Choose a task with an independently checkable result; establish the loop before adding memory or delegation.

Prerequisites: a small task you can describe, its permitted effects, and basic Python for executable examples. Reading the guides needs no API key.

Evidence and coverage

The path links the educational examples as they land. The production harness planned by the maintainer is a separate project.

Follow the concept and build path for runnable lessons and the ecosystem map for wider coverage. Return to all journeys.