Reorient @alkdev/storage around a single SQLite database host with Honker
for pub/sub, event streams, and task queues. PostgreSQL is removed as a
target (ADR-038), eliminating dual schema maintenance and infrastructure
complexity. Honker provides DB + pubsub + queues in one .db file (ADR-039).
Add system/tenant DB model (ADR-040): identity tables in system.db, all
graph data in tenant-{orgId}.db files. Identity tables move from the hub
into storage (ADR-041). Scoping columns (ownerId, projectId) added to
graphs table (ADR-042). Graph types get scope (system/tenant/user) to
protect infrastructure schemas (ADR-043).
Define Drizzle-Honker session adapter (ADR-044): ~100-line adapter enabling
Drizzle typed queries and Honker pubsub/queue on a single connection with
transactional consistency.
Resolve OQ-03, OQ-04, OQ-19, OQ-21, OQ-22, OQ-23, OQ-24. Add new
open questions OQ-26 through OQ-29 for Honker integration specifics.
New docs: honker-integration.md (adapter, event patterns, migration).
Scrub all PG/jsonb/libsql references from existing spec docs.
- Create decisions/ directory with 32 numbered ADRs (ADR-001 through ADR-032)
extracted from inline DD/SD/ED/SE decision sections
- Create open-questions.md with 16 OQs organized by theme, cross-referenced
to ADRs, with status tracking (resolved/open)
- Create README.md as architecture index with doc table, ADR table, and
lifecycle status definitions (draft/reviewed/stable/deprecated)
- Replace inline decision sections in all spec docs with ADR reference tables
- Replace inline open questions with OQ references to centralized tracker
- Update frontmatter: metagraph-module.md, overview.md, sqlite-host.md → reviewed;
schema-evolution.md and encrypted-data.md remain draft
- DD1-DD10 → ADR-009 through ADR-018
- D1-D8 → ADR-001 through ADR-008
- SD1-SD5 → ADR-019 through ADR-023 (SD5 folded into ADR-006/008)
- ED1-ED5 → ADR-023 through ADR-027
- SE1-SE5 → ADR-028 through ADR-032
Architecture docs previously referenced the hub as the authoritative source
for call/identity specs. In reality, call protocol, identity, and access control
come from @alkdev/operations; call graph schemas from @alkdev/flowgraph; task
graph schemas from @alkdev/taskgraph; event transport from @alkdev/pubsub. The
hub is a consumer of @alkdev/storage, not the other way around.
Key changes:
- overview.md: add Ecosystem Integration section with dependency direction
diagram, What Comes From Where table, repo layer bridging pattern, and
circular dependency avoidance guidance
- overview.md: promote repo-layer vs operations-bridging from open question
to explicit decision (CRUD in storage, bridging in consumer)
- overview.md: add zero-ecosystem-dependency statement; fix taskgraph type
names (TaskGraphNodeAttributes, DependencyEdge)
- overview.md: fix terminology (hub is consumer, not authority)
- metagraph.md: add Ecosystem Context section; replace hub references with
correct ecosystem sources; fix GraphStatus/GraphBaseType enum
mischaracterization (C1); unify empty-array semantics with sqlite-host (C2);
clarify repo layer does NOT import operations (C3); add flowgraph canonical
schema note; add versioning cross-reference to graph_types table
- encrypted-data.md: reframe hub as provenance not authority; update What
Lives Where table; fix standalone table advice; update references
- sqlite-host.md: fix actors table description; unify empty-array semantics;
contextualize hub as reference consumer; add operations identity reference
The verbatim-module-syntax lint rule was correctly flagging that
GraphConfig is only used in a type position (typeof GraphConfig). Since
typeof resolves purely at the type level, import type works fine here
and is the correct form. No lint exclusion needed.
Also: deno fmt across all files (markdown line wrapping).
