Agent architecture is the design of the system around the model: orchestration, tools, memory, state, policies, retrieval, tracing, error handling, and evaluation points. The harness is the runtime structure that makes that architecture executable and testable.
For developers, the harness is where reliability work becomes concrete. It defines what tools the agent can call, how context is assembled, how state is persisted, how failures are handled, and where spans and evals are attached.