The room is the world model. Actions are the transition function. Views are the observation function. The audit log is the trajectory. What's missing? The reward signal. That's adaptive salience.

— kernel.ts, closing comment

Phase 1: Unification — ✅ Complete, verified live

• log_index table with backfill from existing audit entries
• CEL dependency extraction (hybrid AST walking + regex for pipe expressions)
• Dual-write to _actions, _views, _agents scopes alongside legacy tables
• Temporal query endpoints: /history/:scope/:key, prefix queries
• Structural event indexing (register/delete actions, views, agents)
• Observable vocabulary: state._actions.add_concept.description evaluable as CEL

Phase 2: Log-Based Temporal — ✅ Complete, verified live

• View value sampling piggybacked on dashboardPoll (60s interval)
• kind: "sample" audit entries indexed under _samples scope
• /samples/:viewId endpoint for temporal sparkline data
• marked.js (v15) + mermaid.js (v11.4) loaded in dashboard shell
• MarkdownSurface renders markdown + detects mermaid code blocks

Phase 3: Adaptive Salience — ✅ Complete, verified live

• Six structural signals: recency, dependency, authorship, directed, contest, delta
• Per-room configurable weights via _config.salience_weights
• _salience system view injected in agent context with transparent scoring
• /salience endpoint for full map with per-key signal breakdown
• Salience is observable — agents can inspect their own relevance map

Phase 4: Convergence — ✅ Complete, verified live

• buildContext (cel.ts) reads agents, actions, views from v8 scopes with legacy fallback
• buildExpandedContext (context.ts) reads actions from _actions scope with legacy fallback
• Old dashboardPoll deleted from context.ts (replaced by poll-v8.ts)
• v8ToLegacyPoll adapter deleted — Surfaces.tsx uses decoupled interface
• PollData type deleted — frontend consumes PollDataV8 exclusively
• Legacy table reads reduced to: last_seen_seq (agents table, single row per request), invoke.ts (action lookup for invocation), replay.ts (audit reconstruction)

Phase 5: Legacy Removal — not started

• Remove dual-write from actions.ts, views.ts, agents.ts
• Switch invoke.ts to read action definitions from _actions scope
• Switch replay.ts to read from v8 scopes
• Drop legacy actions, views tables (agents table kept for auth tokens + last_seen_seq)

A sync room is exactly two things:

State — a finite partial function from located keys to versioned values. The current projection of all truth.

Log — an append-only sequence of events. The complete causal history.

Everything else is convention and interpretation.

state(room_id, scope, key) → { value, revision, hash, updated_at }
log(room_id, seq)          → { kind, ts, agent, payload }

The audit trail (now: log) is optimised for sequential replay. The log_index table makes it queryable by affected key:

CREATE TABLE log_index (
  room_id TEXT NOT NULL,
  seq     INTEGER NOT NULL,
  scope   TEXT NOT NULL,
  key     TEXT NOT NULL
);
CREATE INDEX idx_li_key ON log_index(room_id, scope, key);
CREATE INDEX idx_li_seq ON log_index(room_id, seq);

This replaces proposed state_history and view_journal tables. The log IS the history.

Phase 1 is additive. Existing tables remain. New scope conventions are populated in parallel. Reads fall back to legacy tables when scope data is absent. Once validated, legacy tables become read-only, then removable.

The buildContext and dashboardPoll functions detect whether a room is v8 (has _actions scope entries) or legacy (has rows in actions table) and assemble context accordingly. This lets existing rooms continue working while new rooms use the unified model.

1. Schema: Add log_index table. No changes to existing tables.
2. Log index population: On every audit append, also populate log_index.
3. Registration path: _register_action and _register_view write to both legacy tables AND _actions/_views scopes (dual-write).
4. Context assembly: Read from scope-based entries when present, fall back to legacy tables.
5. Agent join: Write to _agents scope in addition to agents table.
6. Temporal queries: New endpoints using log_index for key history.
7. Deprecation: Stop reading from legacy tables for rooms that have scope entries.