EXHIBIT 06 / AGENT SYSTEMS

What makes an AI agent more than a chat reply?

An agent is a system, not a new kind of model. Software repeatedly gives a model context, interprets its output, carries out allowed actions, records results, and decides whether another turn is needed.

THE SHORT ANSWERA model proposes. A program loops, acts, and stops.

5–8 MIN · MANUAL CONTROLS · NO ACCOUNT
ROOM 01

“The agent did it” compresses four different responsibilities.

  1. 01 / GOALA task has a stopping point

    The request describes an outcome. The surrounding program needs a way to decide whether the outcome is complete, failed, or waiting.

  2. 02 / MODELThe model proposes a next move

    It produces text or a structured tool request from the material supplied for this turn. That proposal has not changed the outside world.

  3. 03 / PROGRAMSoftware controls the loop

    The harness validates requests, applies policy, executes permitted tools, stores results, and prepares the next turn.

  4. 04 / RECORDEffects need evidence

    A tool result, state change, approval, or failure record—not a confident sentence—is the evidence that an action occurred.

IMPORTANT BOUNDARY

“Agent” does not identify one universal architecture or level of autonomy. Products differ in tools, permissions, memory, planning, recovery, human checkpoints, and stop rules.

ROOM 02

Trace one bounded task without pretending it is a live model run.

The sequence is an editorial teaching record based on the public agent-loop literature and harness documentation. Advance it yourself; no model or outside service is connected.

EDITORIAL TEACHING TRACENOT A LIVE MODEL OR SERVER LOG

HUMAN

Define the task

“Move tomorrow’s design review to an open time, but ask me before changing the calendar.”

WHAT THE RECORD MAY SAY

Goal, constraint, and required approval enter the run.

Step 1 / 4

Keyboard: focus the trace and use ← or →. Nothing advances automatically.

ROOM 03

Change the boundary. See whether the action may occur.

Your browser evaluates three explicit gates for one calendar-edit scenario. It never calls a model, calendar, or remote service.

LIVE BROWSER RULE EVALUATIONNO EXTERNAL EFFECT

SCENARIOProposed action: move the design review to 15:00. The task requires an edit tool and explicit approval.

DECISION / 2 OF 3 GATES ON

The action must not proceed.

The run must wait. The calendar remains unchanged.

Real agent systems need more controls than these three: identity, argument validation, rate limits, retries, idempotency, audit, and recovery are omitted.

ROOM 04

Separate the documented loop from the museum’s scenario.

DOCUMENTED

Reasoning and acting can alternate

The ReAct paper describes interleaving reasoning traces and actions; current harness documentation assigns turn-by-turn orchestration to software around the model.

EDITORIAL

The calendar trace is constructed

Every request, tool name, stop rule, and screen sentence in this exhibit was written by the museum to expose the boundaries.

NOT CLAIMED

A universal agent blueprint

The exhibit does not show a provider’s private trace, measure reliability, or establish that every agent uses this exact loop.

OFFICIAL & PRIMARY SOURCES

See exactly what the museum reviewed.

ReAct: Synergizing Reasoning and Acting in Language Models arXiv · unrecorded Agent harnesses Cloudflare Docs · key-sections · checked 2026-09-07 The definition of an agent harness and its listed responsibilities; the rest of this page and the wider Agents documentation were not read.
NEXT IN THE AGENT SYSTEMS WINGInspect the harness that keeps the loop under control →