Skip to content

tcg-platform Current Gap Analysis

This is the current migration gap view. Old phase plans are not canonical.

Capability Current production system tcg-platform status
Customer Telegram Mini App shop exzen-core + Vercel + SQLite Telegram-aware storefront code exists, but production parity is unproven.
Operator Telegram bot exzen-core Not replaced.
Manual/Telegram order intake exzen-core + Google Orders Manual-order capability exists, but no proven production-equivalent Telegram/operator path.
Shopee order ingestion exzen-core; marketplace truth Strong connector/shadow capability, not sole writer yet.
Lazada order/inventory parity exzen-core + Otto Connector code/profiles exist, but live production parity/cutover incomplete.
Inventory authority Google Imports/Inventory Opening snapshot imported; ongoing cutover and reconciliation remain.
Marketplace stock push exzen-core Implemented but disabled to avoid dual writers.
Waybill/physical printing Otto Print-job concepts exist, but Otto is not integrated/replaced.
Direct SPX for Telegram orders exzen-core Not replaced end to end.
Checkout/payments/refunds/returns exzen-core, marketplaces, manual operations Domain pieces exist, full production lifecycle incomplete.
Bank-email transaction ingestion n8n → Sheets No durable replacement inbox/ingest path.
Month-end accounting parity Google finance model Code exists, but real-month reconciliation remains outstanding.
Historical order migration Sheets/SQLite/marketplaces No complete cross-source historical importer/control-total process on main.
General audit ledger Distributed logs/messages/Sheet history Scoped audit events exist; no universal audit trail.
Durable background execution Apps Script, n8n, local schedulers Some platform operations still use detached in-process work rather than durable queues.

Cutover rule

Inventory stock sync is the highest-risk overlap. Keep tcg-platform marketplace stock sync disabled until exzen-core stock sync stops and reconciliation passes.