CORE SOFTWARE SKILLS — LEARNER WORKSHEET
Keep the supplied reference unchanged. Work in your own copy. Predict before executing.
1 Objects: test zero/True/-1/'12000'; explain annotations versus runtime validation.
2 Composition: create two ledgers; add to one; confirm no shared state. Mutate a returned summary.
3 Structures: build an ID list, channel set, checked ID dictionary and FIFO deque. Explain information loss.
4 Complexity: predict list/set counts for 0,1,20,200; state assumptions and memory costs.
5 Package: run CLI from software-kit and another directory; capture stdout/stderr/exit status.
6 REST: start the actual loopback service, record port/status/headers and four-row control.
7 Boundaries: add HTTP tests for boolean amount and repeated JSON keys. Explain 422 versus 400.
8 CI: run compileall, tests and CLI locally. In a copy use a deliberately wrong expected total.
Record GitHub workflow as unexecuted unless you actually run it in your own optional repository.
9 Capstone: add a NEW fifth-order fixture F05/store/2000. Preserve original expected.json.
Add positive and duplicate tests. Write a README and a 150-word evidence-based handover.

EVIDENCE TABLE
Task | prediction | exact command | actual result | pass/fail/unexecuted | limitation
Do not replace a failed expectation with the program output unless the source contract justifies it.
See capstone-rubric.txt; attempt all exercises before opening solutions/worked-solutions.txt.
