CSV Config Pipeline (Archived)

Archived: The admin app now provides CRUD UIs for operational config (topics, seed controls, news config, feature flags, progression). The remaining gaps (model registry, content rules, taxonomy rules) will be addressed by adding admin CRUD pages backed by database storage — not a CSV intermediate layer. See the Admin Config CRUD project for the replacement approach.

Replace hardcoded TypeScript constants and YAML-in-markdown control files with a CSV -> JSON pipeline, pre-populated with all existing values, with round-trip verification.

Key Deliverables

  • Shared CSV reader + validator library
  • CSV joiner for companion files
  • Per-domain export scripts (categories, models, challenge, taxonomy, seed-controls, app-controls)
  • Per-domain sync scripts (CSV -> JSON)
  • Thin TypeScript wrappers importing from generated JSON
  • Round-trip verification (export -> sync -> verify lossless)
  • CI integration

Architecture

Individual sync scripts per domain share a common CSV parsing/validation/joining library. Existing TypeScript files become thin wrappers importing from generated JSON. Export scripts extract current constants to CSV. Verification proves the round-trip is lossless.

Reference Documents