Teaching plan
Preparation and learning goals
Require Exercise 01 or equivalent dependency-remediation knowledge. Check that students can interpret a finding, justify a version change and verify it. Review DEPENDENCY-BASICS.md if needed; first-time dependency learners should complete 01 first.
Select and preflight one manager's kit. Use an institution-approved assistant only; the authored sample is an equal-credit route and does not require skill installation or an account. Actual assistant setup may need separate preparation time. The objective is evidence-based critique and measured verification, not prompting style or provider familiarity.
Suggested sequence
Allow about 30 minutes as a follow-on to 01, or 45–60 minutes for prepared students new to this activity. Timings exclude initial installation. With a live assistant, reserve time for waiting or use the supplied sample.
| Stage | Follow-on time | Checkpoint |
|---|---|---|
| Open evidence and choose actual/sample advice | 5 min | State provenance and protect private data. |
| Critique at least four claims | 10 min | Complete claim/evidence/decision rows, including corrected action. |
| Apply reviewed repair, test and rescan | 10 min | Record measured results, not the assistant's assertions. |
| Reflect | 5 min | Explain limitations and what must not be shared externally. |
For the longer route, add time to orient students in the report and complete verification/reflection. If actual advice contains fewer than four substantive claims, supplement with the labeled authored sample.
Assessment and fallback
This exercise is graded independently out of 20 using its own rubric. Exercise01 knowledge is a prerequisite, not another required submission. Use the dedicated critique table, proposed repair table, before/after verification and privacy reflection.
Sample and actual-assistant routes earn the same maximum. The instructor sample-response-critique.md supplies debrief answers. Historical reference reports support an analysis fallback; clearly distinguish unexecuted repairs/scans from observed results. Never grade provider wording or require fabricated success.