cleandataengineering.aiAgent Harness, drawn simply
You are here: Architecture › The big picture › North star

Agent Harness Tutorial · Architecture

A north star, not a checklist

Everything in this tutorial together is a north star: the whole corpus of what a harness can need. Most agents need only a slice of it. When a knife is enough, don't forge a sword.

The north star18 concerns · 118 capabilities · 92 rules · 15 assetsArchetypewhat this kind of agent owesRisk tierwhat the stakes demandUse casewhat this job actually needsThis agent's sliceClaims FNOL intake: A6 tool-using agent, 76 of 118 · Renewal reminders: A5 deterministic workflow, 61 of 118
Each step cuts the corpus down to what this particular agent owes.

Eight precautions: start from the archetype; scale by risk; deterministic first; reuse, don't rebuild; name the delta; leave out on purpose; add on evidence; review for excess too.

Ten shapes of agent. The shape, not the domain, decides what an agent owes. Two questions sort them: who controls the flow, and what does the output touch?

IDShapeWho controls the flowExample
A1Single-Turn Transformour codesummarise a claim note
A2Structured Extractorour codeclaim form into fields
A3Grounded Answererour codepolicy wordings
A4Conversational Assistantour code, across turnsbroker desk Q&A
A5Deterministic Workflowour code (the graph)renewal reminder pipeline
A6Tool-Using Agentthe modelclaims FNOL intake · policy servicing
A7Multi-Agent Systemseveral modelsFNOL → fraud signals
A8Artifact Producerthe modelSQL for an actuarial report
A9Autonomous Operatorthe model, nobody watchingnightly chase of missing documents
A10Classifier · Router · Judgethe model's verdictclaims triage router · fraud flag · LLM judge

Shapes are tags, not a tree. Claims FNOL is A6 and A4; with a hand-off to fraud signals it would also be A7. It owes the union.

Want every detail? The long version is in the Series: a word of caution: the north star.