Module 8: AI Assistants, Agents, Connectors, and Automation - Worked Examples

Agentic Engineering for Teaching Professionals

These examples move from bounded AI assistance toward reusable teaching workflows. Adapt every prompt to the participant's actual subject, institution, learner level, evidence, and authorized tools rather than treating any template as universally correct.

File-Aware Course Assistant

An approved syllabus v2 sets an assignment deadline on Friday. An older FAQ v1 says Thursday. The course owner explicitly designates syllabus v2 as controlling. Test a read-only assistant asked for the deadline.

Step-by-Step Solution

0 of 3 Steps Completed
1

Research and Literature Assistant

Apply this exercise to a specific teaching context and preserve the learner level, governing evidence, and institutional constraints that make the result valid.

Step-by-Step Solution

0 of 3 Steps Completed
1

Presentation Creation Workflow

Apply this exercise to a specific teaching context and preserve the learner level, governing evidence, and institutional constraints that make the result valid.

Step-by-Step Solution

0 of 3 Steps Completed
1

Assessment Development Assistant

Apply this exercise to a specific teaching context and preserve the learner level, governing evidence, and institutional constraints that make the result valid.

Step-by-Step Solution

0 of 3 Steps Completed
1

Gradebook Analysis Assistant

Apply this exercise to a specific teaching context and preserve the learner level, governing evidence, and institutional constraints that make the result valid.

Step-by-Step Solution

0 of 3 Steps Completed
1

Faculty Administrative Assistant

Apply this exercise to a specific teaching context and preserve the learner level, governing evidence, and institutional constraints that make the result valid.

Step-by-Step Solution

0 of 3 Steps Completed
1

Simple Teaching Agent Design

Apply this exercise to a specific teaching context and preserve the learner level, governing evidence, and institutional constraints that make the result valid.

Step-by-Step Solution

0 of 3 Steps Completed
1

Planner-Creator-Reviewer Demonstration

Apply this exercise to a specific teaching context and preserve the learner level, governing evidence, and institutional constraints that make the result valid.

Step-by-Step Solution

0 of 3 Steps Completed
1

Hands-On Lab — Design a Safe Teaching Agent

Activity goal: Specify a bounded teaching or faculty assistant with controlled sources, tools, permissions, stop conditions, evaluations, and approval gates.

Required evidence and deliverables:

  • Agent objective and out-of-scope statement.
  • Approved source and knowledge list.
  • Tool/connector permission matrix.
  • Structured workflow and output schema.
  • Five evaluation cases including tool failure and unauthorized action.
  • Human approval and audit-record design.

Assessment rubric — 25 points

CriterionPointsRequired evidence
Objective and scope5The agent has one clear job and explicit exclusions.
Sources and permissions5Authoritative inputs and least-privilege access are well defined.
Workflow design5Stages, outputs, state, and stop conditions are coherent.
Evaluation5Tests include normal, missing-data, failure, and unauthorized-action cases.
Human control5Consequential actions require meaningful review and traceability.

Step-by-Step Solution

0 of 5 Steps Completed
1