Quality fixes:
- Add deprecation warnings to 5 silent no-op legacy wrappers in assess.py
- Remove dead code in tracker.py score_by_section (unused first loop)
- Remove unused variable in assess.py _check_order
- Fix .gitignore: add progress JSONs, checkpoints, .coverage, .DS_Store, LaTeX aux
- Fix "all three plans" → "all four plans" in learning_objectives.md
- Add teaching/ package to README project tree
- Add compendium to README paper tree
Testing:
- Add 43 unit tests for teaching/assess.py and tracker.py (quiz, predict_choice,
reflect, order, checkpoint_summary, legacy wrapper deprecation warnings,
tracker scoring, persistence, mastery calculation)
- Add notebook execution test suite (nbclient): all 11 notebooks execute without
errors in a fresh kernel, structural validation (valid JSON, has code cells,
has assessments, section parameters, learning objectives document)
- Overall test count: 185 passing (was 107), coverage: 85% (was ~25% in tests)
Toolchain:
- Add pytest-cov, ruff, nbclient, nbformat to dev dependencies
- Add ruff config (E, F, W, I, UP, B, SIM rules)
- Add coverage config with term-missing output
- Fix all ruff lint issues across src/ and tests/ (import sorting, unused imports)
- Fix Plan D notebook paths (configs/rungs → ../../configs/rungs)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Plan D structures the material as a scientific argument chain:
- Experiment 1: Can the [[4,2,2]] code protect a magic state? (W=1.0, 12/12 errors)
- Experiment 2: How much magic survives noise? (scoring, parameter sweeps)
- Experiment 3: Can a ratchet learn to optimise? (monotonic improvement, transfer)
Each notebook follows Hypothesis → Claim → Experiment → Proof → Next Hypothesis.
Includes builder script, updated learning objectives, README, and compendium cross-ref.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
