* docs(memory): add memory provenance and deletion concepts page
New concepts/memory-provenance page tells the operator policy story end to end: recorded session lineage through consolidation, the admission policy and its pipeline-only boundary, memory forget guarantees (whole-entry purge, derived-artifact sweep, forgotten tombstones) and reported boundaries (transcripts, curated writes, paraphrased prose), plus the person/source purge workflow. Cross-linked from memory architecture, dreaming, builtin, overview, CLI, and config reference; registered in docs navigation.
* docs(i18n): add glossary terms for the memory provenance page
* docs(memory): distinguish curated entries from missing lineage in the provenance page
* docs(memory): document shipped turn-taint propagation for network tool output
docs/concepts/memory-architecture.md still claimed content origin is not
propagated within an owner turn. The runtime has since shipped exactly that:
tool results declaring resultContentSource "network" taint the rest of the
turn (packages/agent-core/src/agent-loop.ts), tainted assistant messages
classify untrusted for memory (packages/memory-host-sdk session-provenance),
and taint clears on the next user message. The remaining gap - tools that do
not declare a network content source do not taint - is now stated honestly.
* docs(memory): state untainted assistant provenance as agent, not sender class
* feat(memory): add provenance and recall metadata to the memory index
* feat(memory): provenance-gated promotion and capture hygiene
* feat(dreaming): LLM consolidation with deterministic gates, on by default
* feat(active-memory): deterministic recall lane with escalation default
* feat(memory): user model file and standing intents
* docs(memory): document the memory architecture
* fix(memory): live-QA fixes — metadata writers, provenance classes, intent scope, claim accumulation
Peter Steinberger