Skip to content

Scenario Selection Guide

Guide Maps

graph LR
  shape["Public shape routes"] --> definition["Definition-time review"]
  definition --> invoke["Invocation routes"]
  invoke --> proof["Proof routes"]
flowchart LR
  question["What do you need to learn or prove?"] --> route["Choose the matching public route"]
  route --> guide["Open the matching local guide"]
  guide --> proof["Escalate to tests or saved bundles only when needed"]

Use this guide when the capstone now has enough routes that choosing the next one could become its own source of noise. The goal is to keep route selection proportional to the metaprogramming question you actually have.

Route matrix

If the question is about... Start with Then
what exists publicly before invocation make manifest, make registry, make plugin PLUGIN_RUNTIME_GUIDE.md and INSPECTION_GUIDE.md
what generated call shapes look like make signatures DEFINITION_TIME_GUIDE.md and tests/test_registry.py
what one concrete action does make demo SCENARIO_GUIDE.md and PLUGIN_CATALOG.md
what the wrapper adds at runtime make trace TRACE_GUIDE.md and tests/test_runtime.py
what can be reviewed later without rerunning commands make inspect, make tour, or make verify-report BUNDLE_GUIDE.md and PROOF_GUIDE.md
what the strongest local bar is make confirm TEST_GUIDE.md

Best entry by learner need

  • start with the public shape routes if metaclass and descriptor consequences still feel abstract
  • start with signatures if generated constructors are the main source of confusion
  • start with trace if decorator behavior still feels magical
  • start with saved bundles if the individual commands are clear but the whole review route is not

Best companion guides