AC
Aigarth Tracker
All phases
05

ANN Platform

Registry, builder, versioning, marketplace, licensing, analytics.

Progress
100%
Tasks done
21 / 21
0 remaining
Owner
ANN lead
Status: complete
Tasks
Kanban for this phase. Move tasks between columns to update status.
Backlog
0
To do
0
In progress
0
In review
0
Done
21
ANN service scaffold (Fastify + Drizzle + node-postgres + JWT, port 7006)
p0
Eng lead + AI agents1 SP
ANN schema (9 tables: categories, licenses, anns, versions, deployments, ratings, benchmarks, license_grants, audit log)
p0
Eng lead + AI agents3 SP
Categories (16) + Licenses (4) seed, upsert-aware
p0
Eng lead + AI agents1 SP
ANNs service: create draft, update, publish, deprecate, sign, list/search/filter
p0
Eng lead + AI agents3 SP
Versions service (semver, immutable, latest-flip + denormalized metrics)
p0
Eng lead + AI agents2 SP
Deployments service (forwards to services/compute with stub fallback)
p0
Eng lead + AI agents2.5 SP
Ratings (with denormalized aggregates) + Benchmarks
p0
Eng lead + AI agents2 SP
License grants service (caller-side grant + revoke)
p0
Eng lead + AI agents1 SP
23 endpoints across /v1/anns, /v1/categories, /v1/licenses, /v1/me
p0
Eng lead + AI agents2.5 SP
Shared serializers in lib/serialize.ts (new convention)
p1
Eng lead + AI agents0.5 SP
Audit log (ann_audit_logs, free-form action text)
p1
Eng lead + AI agents0.5 SP
E2E test: 10 sections, ~40 assertions, all green
p0
Eng lead + AI agents2 SP
Phase 5 delivery report with time-to-ship metrics
p0
Eng lead + AI agents2 SP
Gap 1: Deploy transactional (fail closed)
p0
Eng lead + AI agents1 SP
Gap 2: Payment-gated license grants (Stripe-style)
p0
Eng lead + AI agents2 SP
Gap 3: Admin endpoints (suspend, restore, force-deprecate)
p0
Eng lead + AI agents2 SP
Gap 4: Full-text search with pg_trgm + GIN index
p0
Eng lead + AI agents1.5 SP
Gap 5: Rate limiting on write endpoints (global preHandler)
p0
Eng lead + AI agents1 SP
Gap 6: Idempotency on POST /v1/anns (DB-backed)
p0
Eng lead + AI agents2 SP
Gap 7: Analytics endpoint (deployments + rating distribution + benchmarks)
p1
Eng lead + AI agents1.5 SP
E2E extended: 15 sections (was 10), all green
p0
Eng lead + AI agents1 SP