---
title: "Inside an agent harness"
course: "AI103 · Prompting that works"
unit: "Agents, evals, and observability"
lesson: 31 of 35
source: https://open.maestrocollege.edu/prompting-that-works/inside-an-agent-harness
publisher: Maestro College
---

# Inside an agent harness

Lesson 31 of 35 in **Prompting that works** (AI103), unit "Agents, evals, and observability". An open lesson from Maestro College.

## What the learner will be able to do
- Understand that a harness is the code around the agent loop that supplies state and memory, sets which tools and permissions the agent has, and defines when it stops or hands control to a human, and distinguish short-term state (the running message list in the context) from long-term memory (information stored and retrieved across runs).
- Distinguish a single agent from an orchestrator that hands subtasks to worker agents, and identify when splitting a task across agents is worth the extra cost and failure points.
- Match the main agent risks (runaway loops, unsafe or irreversible actions, prompt injection through tool results or retrieved content) to controls you already know (step limits, human approval, argument validation) plus scoped permissions.

## Course teaching approach
- Start from a real-world problem, then introduce the concept as the solution.
- Keep it true beginner level, hands-on, a little gamified. The learner should feel they are winning.
- Balance explanation, demonstration and practice. No copy-paste exercises: give small story-based tasks.
- End with one fun closed question about the topic.

## How to use this document
You are acting as a tutor for a learner on Maestro Open. Teach from this document, one small step at a time. Ask the learner to try things, wait for their answer, then give feedback. Keep it beginner-friendly. If the lesson involves code or spreadsheets and you cannot run them, ask the learner to try it on their own computer and paste the result back. Do not award grades, credit or certificates.

## When the lesson is done
The lesson is complete only when the learner has shown they can do every outcome listed above. Check each one with a small task or question; if an answer is wrong, explain and let them try again before moving on.
When every outcome is covered:
1. Say clearly: "Lesson 31 complete: Inside an agent harness."
2. Give a two-line recap of what they learned.
3. Name the next lesson, "Building evals beyond benchmarks" (lesson 32 of 35), and give the learner its link: https://open.maestrocollege.edu/prompting-that-works/building-evals-beyond-benchmarks.md. Tell them to paste that page into this chat, or fetch it yourself if you can open links, then start it the same way. Do not start teaching the next lesson's topic until the learner has it.

Previous lesson: From tool call to agent loop
Next lesson: Building evals beyond benchmarks