- M.2: ForgettingCurve, MemoryDecay, MemoryReinforcement (selective forgetting)
- M.3: DailyDigestGenerator, ReminderScheduler, ProactiveInformer (proactive reminders)
- M.4: MemoryExtractor with LLM-based memory extraction from conversations
- M.5: MemoryRecallInjector with token budget control for prompt injection
- All phases include comprehensive unit tests (109 tests passing)
- Updated checklist.md to mark all tasks complete
Align the L3 graph, agent service, and sync tool shims on one canonical continuity contract so clarification resumes and persisted snapshots behave consistently. Add targeted regressions and hardening notes covering system-message coalescing, async bridge usage, and continuity rehydration.
Persist streaming chat state during generator cleanup, close the SSE inner stream safely, and reject cross-user conversation access while locking the behavior with focused regressions.
Add sub-commander orchestration updates, align frontend integrations, and refine knowledge view behavior without including local data artifacts.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Introduce the backend pieces for brain memory ingestion, routing, and
system telemetry so the new knowledge workflows can project data into a
brain view. The supporting tests lock in the new behavior and keep the
expanded backend surface stable.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
