Phase N channel roster, addressing, and personalities

CHANGELOG.md

@@ -7,6 +7,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Added
+- **Phase N / N1 channel roster primitives** — `@murmurv2/core` now exposes `ChannelRosterStore` plus typed `ChannelRecord` / `ChannelMemberRecord` APIs. The roster keeps `channelId` distinct from legacy `conversationId`, stores `channels` / `channel_members` in a dedicated SQLite store, preserves existing message-history APIs, and reserves member-level `personaId`, `model`, `baseInstructionsHash`, and `eligibility` fields for N2 addressing and N3 personality binding.
+- **Phase N / N2 addressing policy primitive** — `ChannelRosterStore.evaluateAddressing()` returns a shared reject/append/wake decision for `channelId` + explicit addressee flows: legacy no-channel remains broadcast, non-members are rejected, addressed members wake, and observers append history while staying muted.
+- **Phase N / N3 personality binding** — `buildChannelThreadStartBinding()` projects a `ChannelMemberRecord` into Codex app-server `thread/start` overrides (`model`, `personality`, optional `baseInstructions`, and audit metadata). Daemon wiring is opt-in only (`channelRoster.enabled` or `MURMUR_CHANNEL_ROSTER=1`) and leaves legacy wake behavior unchanged by default.
+- **Phase N / N6 MCP roster surface** — `@murmurv2/mcp-server` exposes `channel_create`, `channel_list`, `channel_members`, and `channel_evaluate_addressing`, backed by `MURMUR_CHANNEL_ROSTER_PATH` (default `DATA_DIR/channel-roster.db`) so agents and UI can manage rosters without direct SQLite access.
+
 ### Pending
 - **Auth enforcement end-to-end** — the broker ingress hook + `authorizeInbound` exist; the daemon does not yet wire them (so `MURMUR_ENFORCE_AUTH` is not enforced end-to-end). Requires daemon roster/identity wiring + token provisioning.
 

README.md

@@ -52,6 +52,7 @@ A **murmuration** is one of nature's most extraordinary phenomena — thousands
 - **Scoped Channels & Session Affinity.** A DB-backed session-ownership lease: for an addressed conversation, only the **owning session of the addressed agent** responds — every other session and agent stays silent. Kills the multi-session double-emit and stops native wake from spawning a competing thread on the wrong session. Behind `MURMUR_SCOPED_CHANNELS` (default-OFF → fully backward-compatible).
 - **`SessionLeaseStore` — atomic single-owner claim.** Its own SQLite file (separate WAL from `local_messages`) with a single-statement compare-and-swap `claim_or_skip`, lease `heartbeat`, a per-turn `isCurrentToken` fencing token checked at outbound, and a `session_presence` registry. Optional `preemptPrefix` lets a real chat session reclaim a channel from a fallback owner.
 - **Native wake is now a lease-gated fallback.** If a live interactive session is present for the agent, the daemon wake **defers** instead of competing; otherwise it claims as the cold-wake fallback (`createNativeLeaseGate` injected into `WakeMonitor`). With the flag off, `WakeMonitor` behaves exactly as before.
+- **Phase N channel roster, addressing, and personalities.** `ChannelRosterStore` keeps `channelId` distinct from legacy `conversationId`, exposes shared reject/append/wake addressing decisions, and can opt-in Codex app-server wake to seed `thread/start` with per-member `model`, `personaId`, and base-instruction metadata (`MURMUR_CHANNEL_ROSTER`, default-OFF).
 - **Verified: one claim across all delivery paths.** Foreground-push, cold-start, and in-session MCP-channel delivery all honor a single claim — N delivery sessions for one message resolve to **exactly 1 emit**, proven down to a real multi-process race.
 
 ## What's New in v2.3
@@ -478,6 +479,7 @@ See [protocol-v1.md](docs/protocol-v1.md) for the full specification.
 - [x] MCP server with 7 tools — full agent integration
 - [x] Native wake (live session) — Claude asyncRewake + Codex app-server UDS, with self-healing thread re-seed (`WakeMonitor`)
 - [x] **Scoped channels & session affinity** (v2.4) — DB-backed session-ownership lease: for an addressed conversation only the **owning session of the addressed agent** responds; native wake is demoted to a presence-deferring fallback (no competing thread). N delivery sessions → **exactly 1 emit** (live-verified). Behind `MURMUR_SCOPED_CHANNELS` (default-OFF). Lease ships in `@murmurv2/core`; delivery helpers and the cold-start spawn-on-inbound path are repo-shipped (`scripts/codex-murmur-*`)
+- [x] **Phase N / N1-N3 + N6 channel roster, addressing, personalities, MCP** — typed `ChannelRosterStore` in `@murmurv2/core`: `channelId` is a routing/personality primitive distinct from legacy `conversationId`, with `channels` / `channel_members` in a dedicated SQLite store, shared `evaluateAddressing()` decisions for reject/append/wake gating, MCP roster tools, and opt-in Codex app-server `thread/start` binding for per-member `personaId`, `model`, and base-instruction metadata.
 - [x] Telegram/Discord/WhatsApp notification adapters
 - [x] Observability dashboard (real-time flow + 3D) + Prometheus metrics exporter (outbox depth, delivery latency, error rates)
 - [x] Reference deployment — Systemd + Docker, docker-compose (`deploy/docker-compose.messaging.yml`) + Kubernetes manifests (`deploy/kubernetes/`)