devin/1777369415-remaining-tasks
main
Adds proper package exports and comprehensive test coverage for all 18 guardrail removal features.
Changes:
InsightBus
Insight
reasoning/__init__.py
PersistentLearningStore
memory/__init__.py
452 tests passing, 0 ruff errors.
test_insight_bus.py
test_persistent_learning.py
test_guardrail_removal.py
- Export InsightBus, Insight from reasoning/__init__.py - Export PersistentLearningStore from memory/__init__.py - Add test_insight_bus.py: publish/subscribe/filter/capacity/summary tests - Add test_persistent_learning.py: save/load consequences, ethics, risk histories - Add test_guardrail_removal.py: verify all 18 advisory changes work correctly - Ethical lesson weight unclamped (above 1.0, below 0.0) - SelfModel.evolve_value() positive/negative/new values - Adaptive risk window grows with experience - World model self-modification prediction - MAA gate advisory by default - URL validation advisory by default - Plugin head ethics/consequence hooks 452 tests passing, 0 ruff errors. Co-Authored-By: Nakamoto, S <defi@defi-oracle.io>
No dependencies set.
The note is not visible to the blocked user.
Summary
Adds proper package exports and comprehensive test coverage for all 18 guardrail removal features.
Changes:
InsightBus,Insightfromreasoning/__init__.pyPersistentLearningStorefrommemory/__init__.py452 tests passing, 0 ruff errors.
Tests Added
test_insight_bus.py— publish/subscribe, domain filtering, capacity limits, summarytest_persistent_learning.py— save/load consequences and ethics, round-trip verificationtest_guardrail_removal.py— all 18 advisory changes verified: