Backend implementation tasks
Status date: 2026-07-24
Phase 1 - isolated backend baseline
- Create
E:\workspace\server\saas-voc-server as an independent Git repository.
- Keep
moshengqi-server and future-server read-only.
- Add strict environment validation without credential fallbacks.
- Add pooled PostgreSQL access and an idempotent migration runner.
- Add an independent Parse Server application configuration.
- Add the initial domestic VOC schema, constraints, foreign keys, and indexes.
- Seed the
demashi workspace and JD source metadata.
- Add health, sync, job-status, and snapshot endpoints.
- Add a database-free local-demo server for frontend integration using the normalized Demashi snapshot.
- Add Fmode timeout/retry/authentication behavior behind a server-only client.
- Add unit tests and compile under strict TypeScript settings.
- Add a shared platform repository contract with local-memory and PostgreSQL implementations.
- Add Parse-session authentication, workspace membership, and four-role RBAC.
- Add cursor-paginated product, relation, review, job, import, and platform workflow APIs.
- Add analysis requests, action workflows, alerts, and audit history without fake AI results.
- Cover local platform flows, viewer write denial, Parse session behavior, pagination, and audit writes.
- Make the database-free runtime reusable through a configurable local workspace id.
- Reject partial platform migrations through the health check.
- Add manual sync retry/cancel, local job events, and stale-worker job recovery.
- Protect owner membership and validate action assignees inside the workspace boundary.
- Run migrations against a newly provisioned PostgreSQL database.
- Smoke-test the local HTTP service under the supported Node 22 runtime.
- Add a Parse REST storage driver that preserves the existing frontend/backend API contract.
- Add isolated master-key-only
Voc* schemas and remote schema reconciliation.
Phase 2 - case-data persistence
Phase 3 - real JD source contract
Phase 4 - deployable closure
Explicitly deferred
- Multi-platform adapters beyond JD.
- Multi-tenant billing, SSO provisioning, and custom roles beyond the four built-in roles.
- Scheduled full-catalog crawling.
- AI reports, sentiment, pain points, or recommendations without review evidence.
- Any reuse of the cross-border production database or credentials.